Upload
others
View
2
Download
0
Embed Size (px)
Citation preview
EARLY-LIFE VIRAL INFECTION IMPACTS BRAIN DEVELOPMENT IN THE PIGLET
BY
MATTHEW SCHREIBER CONRAD
DISSERTATION
Submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy in Neuroscience
in the Graduate College of the University of Illinois at Urbana-Champaign, 2014
Urbana, Illinois Doctoral Committee: Professor Rodney W. Johnson, Chair, Director of Research Professor Janice M. Juraska Associate Professor Justin S. Rhodes Associate Professor Bradley P. Sutton
ii
ABSTRACT
During infancy and childhood, almost every person will impacted by infection. Neonates,
due to their immature immune system, are especially vulnerable to respiratory infection and are
more prone to hospitalization. Peripheral immune activation in adulthood has been shown to
have neurobehavioral effects, but little is known about how it can impact brain development.
Gliogenesis, synaptogenesis, and myelination peak during the neonatal period and are all prone
to environmental insults, including infection. Therefore, the broad aim of this research was to
characterize the impact of early-life respiratory viral infection on brain growth and development.
Previous work in our lab has shown that porcine reproductive and respiratory syndrome
virus (PRRSV) causes microglia activation, neuroinflammation, and cognitive deficits in piglets.
This work further expands on these findings to look at changes to brain development from a
macro and microscopic perspective. In order to non-invasively track macro changes to brain
development, magnetic resonance imaging (MRI) procedures were developed to measure brain
region volumes in the piglet. Using these methods, the normal brain development of the
domestic pig was characterized in order to compare brain growth trajectories to what is known
for humans. Additionally, advanced analysis methods for determining gray and white matter
changes using voxel based morphometry is now possible due to creation of an averaged piglet
brain and MRI-based atlas. This method, along with diffusion tensor imaging (DTI) and
magnetic resonance spectroscopy (MR-Spec), allow for in vivo quantification of brain growth
and development.
Using these imaging methods, the impact of neonatal PRRSV infection was
characterized. Reductions in gray and white matter in the primary visual cortex were found in
the PRRSV piglets. Additionally, there was a trend for a decrease in fractional anisotropy in the
corpus callosum suggesting either delayed or disrupted myelination in the PRRSV piglets. MR-
Spec analysis revealed a significant change in three metabolites in the hippocampus. N-
iii
acetylaspartate, creatine, and myo-inositol were all decreased in PRRSV piglets
showing that infection caused an energy imbalance and may have impacted neuron and
astrocyte health. Collectively, these data show that early-life infection can affect grey
and white matter development and cause metabolite disturbances.
In addition to the regional changes to brain development above, changes at the
cellular level were explored. Since PRRSV causes deficits in hippocampal-dependent
learning and memory, changes to neurogenesis and neuron morphology were assessed
as they have been shown to contribute to hippocampal learning. A sexual dimorphism
was found in the number of surviving newly divided cells (BrdU+) where males had
increased levels compared to females. PRRSV infection caused a significant reduction
in males only. Cell fate analysis showed that 80% of the newly divided cells form
neurons in controls and this is reduced to 57% in PRRSV piglets. Fifteen percent of the
newly formed cells become microglia and this is stable across sex and PRRSV
infection. A sexual dimorphism was also found in dentate granule neuron morphology
where males had more complex dendritic trees than females in the outer cell layer.
PRRSV infection caused a change in the dendritic tree shape where the initial branch
point was further away from the cell soma in the inner granule cell layer. No differences
were found in dendritic spine density. Taken together, PRRSV causes changes to
neurogenesis and neuron morphology, which may contribute to the deficits in learning
and memory.
In order to mitigate the changes to brain development caused by PRRSV,
minocycline was used to prevent microglia activation. Minocycline is a second
generation tetracycline antibiotic which has been shown to reduce microglia activation
iv
and provide protection in rodent models of neuroinflammation. Three weeks of high
dose minocycline administration failed to provide neuroprotection in the PRRSV model.
Minocycline treatment caused increased microglia activation and pro-inflammatory
cytokine production. The high dose of minocycline may have led to increased
intracranial pressure or bilirubin-induced neuroinflammation causing the
neuroinflammation. These findings suggest that high dose chronic minocycline
treatment is not appropriate attenuating microglia activation in neonates.
Together, these novel data show that early-life respiratory viral infection can
impact brain development. Inflammation can disrupt sensitive developmental
processes, some in a sexually dimorphic manner. Long-term studies are needed to see
if these changes are permanent or if the brain is able to recover.
v
ACKNOWLEDGEMENTS
I would like to give many thanks to my doctoral advisor, Dr. Rodney Johnson. He has
given me a tremendous amount of support during my graduate school career. The willingness
to let follow my passions with developing new MRI techniques has allowed me to step out of my
comfort zone and gain a small amount of expertise in a field that always fascinated me. I
appreciate all of the mentoring and the support to travel across the world to present work from
our lab. I would also like to thank Dr. Bradley Sutton for all of his assistance with the MRI work.
I have gained so much knowledge and it will be extremely helpful for my clinical training. I also
owe a debt of gratitude to Dr. Janice Juraska and Dr. Justin Rhodes for their mentoring.
Through discussions of data, study design, and techniques, they have contributed greatly to my
training. I am also very grateful for my undergraduate advisor, Dr. Andrea Neely, who taught
me the basics of scientific research and inspired me to get my MD/PhD.
I would like to extend a special thanks to Dr. Ryan Dilger who took me in under his wing
when I first started in the lab. You have taught me so many things over the years including how
to handle piglets and sows. You were always available to bounce ideas off of and to help with a
problem. Your friendship, both in and out of the workplace, has been incredible.
Also, thank you to all of the other graduate students in the lab. I have learned so much
from you throughout the years and look forward to becoming collaborators in the future. Help
from the undergraduate students was invaluable as it took a small army to conduct these
studies. Also, thank you to my friends in the neuroscience and medical scholars program. I
have met so many great people and look forward to continuing the friendships in the future.
Finally, I have to give a huge thank you to my wife, Marti, and my family. I still think that
it is fate that brought me to Illinois and that I met you my first day being here. Your unwavering
vi
love and support has been my inspiration and motivation. Thank you to my parents, Lauree and
Dan, and brother, Greg, for always supporting me. You still think that going to school for this
long of a time is crazy (and probably right), but you have always supported me.
vii
TABLE OF CONTENTS CHAPTER 1 GENERAL INTRODUCTION AND JUSTIFICATION ............................................. 1
1.1 References ................................................................................................................. 3
CHAPTER 2 LITERATURE REVIEW ........................................................................................... 5 2.1 Neurodevelopment of the Domestic Pig ..................................................................... 5 2.2 Neuroimaging in the Piglet ......................................................................................... 7 2.3 The Piglet as a Model for Human Neurodevelopment ............................................. 10 2.4 Piglet Models in Neonatal Research ........................................................................ 11 2.5 Microglia and the Developing Brain ......................................................................... 14 2.6 Immune to Brain Communication ............................................................................. 17 2.7 Pathogenesis of Porcine Reproductive and Respiratory Syndrome Virus ............... 19 2.8 PRRSV and Neuroinflammation ............................................................................... 20 2.9 Inflammation and Neurogenesis .............................................................................. 22 2.10 Inflammation, Neuron Morphology, and Dendritic Spines ...................................... 24 2.11 Minocycline and Neuroinflammation ...................................................................... 26 2.12 Summary and Objectives ....................................................................................... 27 2.13 References ............................................................................................................. 30
CHAPTER 3 EARLY POSTNATAL RESPIRATORY VIRAL INFECTION INDUCES STRUCTURAL AND NEUROCHEMICAL CHANGES IN THE NEONATAL PIGLET BRAIN ... 42
3.1 Abstract .................................................................................................................... 42 3.2 Introduction .............................................................................................................. 43 3.3 Materials and Methods ............................................................................................. 44 3.4 Results ..................................................................................................................... 50 3.5 Discussion ................................................................................................................ 53 3.6 Figures and Tables .................................................................................................. 58 3.7 References ............................................................................................................... 66
CHAPTER 4 EARLY POSTNATAL RESPIRATORY VIRAL INFECTION ALTERS HIPPOCAMPAL NEUROGENESIS, CELL FATE, AND NEURON MORPHOLOGY IN THE NEONATAL PIGLET .................................................................................................................. 70
4.1 Abstract .................................................................................................................... 70 4.2 Introduction .............................................................................................................. 71 4.3 Materials and Methods ............................................................................................. 72 4.4 Results ..................................................................................................................... 80 4.5 Discussion ................................................................................................................ 83 4.6 Figures ..................................................................................................................... 88 4.7 References ............................................................................................................... 96
CHAPTER 5 MINOCYCLINE ADMINISTRATION DOES NOT ATTENUATE MICROGLIA ACTIVATION AND NEUROINFLAMMATION INDUCED BY PORCINE REPRODUCTIVE AND RESPIRATORY SYNDROME VIRUS ...................................................................................... 100
5.1 Abstract .................................................................................................................. 100 5.2 Introduction ............................................................................................................ 101 5.3 Materials and Methods ........................................................................................... 103 5.4 Results ................................................................................................................... 107 5.5 Discussion .............................................................................................................. 109 5.6 Figures ................................................................................................................... 113 5.7 References ............................................................................................................. 118
viii
CHAPTER 6 SUMMARY AND SIGNIFICANCE...................................................................... 121 APPENDIX A MAGNETIC RESONANCE IMAGING OF THE NEONATAL PIGLET BRAIN . 124
A.1 Abstract .................................................................................................................. 124 A.2 Introduction ............................................................................................................ 125 A.3 Materials and Methods .......................................................................................... 127 A.4 Results ................................................................................................................... 131 A.5 Discussion ............................................................................................................. 132 A.6 Figures and Tables ................................................................................................ 136 A.7 References ............................................................................................................ 141
APPENDIX B BRAIN GROWTH OF THE DOMESTIC PIG (SUS SCROFA) FROM 2 TO 24 WEEKS OF AGE: A LONGITUDINAL MRI STUDY ................................................................ 144
B.1 Abstract .................................................................................................................. 144 B.2 Introduction ............................................................................................................ 145 B.3 Materials and Methods .......................................................................................... 147 B.4 Results ................................................................................................................... 149 B.5 Discussion ............................................................................................................. 151 B.6 Figures and Tables ................................................................................................ 156 B.7 References ............................................................................................................ 161
APPENDIX C AN IN VIVO THREE-DIMENSIONAL MAGNETIC RESONANCE IMAGING-BASED AVERAGED BRAIN AND ATLAS OF THE NEONATAL PIGLET (SUS SCROFA) .. 164
C.1 Abstract ................................................................................................................. 164 C.2 Introduction ............................................................................................................ 165 C.3 Materials and Methods .......................................................................................... 167 C.4 Results and Discussion ......................................................................................... 171 C.5 Figures and Tables ................................................................................................ 174 C.6 References ............................................................................................................ 179
1
CHAPTER 1
GENERAL INTRODUCTION AND JUSTIFICATION
Environmental insults such as malnutrition, traumatic stress, and infection during
sensitive developmental periods can affect neural cells and circuits, slow cognitive
development, and increase the risk for behavioral disorders (Shah, Doyle et al. 2008).
However, despite the high susceptibility of the human neonate to infection and a recent
prospective study suggesting a correlation between neonatal infection and adverse
neurodevelopmental outcomes, the ramifications of postnatal infection on brain and cognitive
development are poorly understood (Stoll, Hansen et al. 2004). The potential effects of
postnatal infection on brain growth and cognitive development are of particular interest because
the human brain undergoes extraordinary growth during the first year of life and disruption of
developmental processes may have long-lasting or permanent effects on brain structure and
function.
Supporting this notion, inflammation has been shown to inhibit neurogenesis (Ekdahl,
Claasen et al. 2003). Furthermore, epidemiological studies have linked childhood infection with
increased risk of schizophrenia, autism, and other affective disorders (Bode, Ferszt et al. 1993;
Rantakallio, Jones et al. 1997; Atladottir, Thorsen et al. 2010). Likewise, maternal infection
during pregnancy as well as infection in preterm infants is a risk factor for neurodevelopmental
deficits and mental illness (Stoll, Hansen et al. 2004; Boksa 2010). Studies in rodents have
shown that adult offspring from infected dams have behavioral abnormalities reminiscent of
those seen in humans with psychiatric illnesses (Fatemi, Reutiman et al. 2008). The effects of
infection on behavioral development are not caused by the pathogen per se, but rather the
host’s inflammatory response against it (Bilbo and Schwarz 2009). Peripheral infection is well
known to activate microglia, which we suspect creates an inflammatory environment that affects
neurodevelopmental processes including neurogenesis. Abnormal immune activation is seen in
2
psychiatric illnesses, including increased circulating proinflammatory cytokines and activated
microglia (van Berckel, Bossong et al. 2008; Ashwood, Krakowiak et al. 2011). In addition to
abnormal cytokine profiles, deficits in learning and memory as well as abnormal hippocampal
morphology have been reported in association with pre- and postnatal infection (Fatemi, Earle
et al. 2002; Bilbo, Rudy et al. 2006). Similar structural changes and abnormal hippocampal
volumes have been reported in schizophrenia and autism disorders (Nelson, Saykin et al. 1998;
Rosoklija, Toomayan et al. 2000).
Thus, understanding how infection affects brain and cognitive development may be key
to preventing and treating mental illness, which affects more than 25 million Americans each
year (Kessler, Chiu et al. 2005). The broad working hypothesis is that neonatal infection affects
brain growth and development, leaving behind a “finger print” that leads to behavioral disorders
later in life. The goal of this research was to determine if neonatal infection results in changes
to multiple levels of brain structure. First, MRI was used to determine regional changes to grey
and white matter development including neurochemical concentration analysis of the
hippocampus. Secondly, focus was placed cellular changes in the hippocampus including
neurogenesis, cell fate, and neuron morphology. Lastly, studies were conducted to test the
potential therapeutic value of minocycline for protection against changes induced by infection.
3
1.1 References Ashwood, P., P. Krakowiak, et al. (2011). "Elevated plasma cytokines in autism spectrum
disorders provide evidence of immune dysfunction and are associated with impaired behavioral outcome." Brain Behav Immun 25(1): 40-45.
Atladottir, H. O., P. Thorsen, et al. (2010). "Association of Hospitalization for Infection in
Childhood With Diagnosis of Autism Spectrum Disorders: A Danish Cohort Study." Arch Pediatr Adolesc Med 164(5): 470-477.
Bilbo, S. D., J. W. Rudy, et al. (2006). "A behavioural characterization of neonatal infection-
facilitated memory impairment in adult rats." Behav Brain Res 169(1): 39-47. Bilbo, S. D. and J. M. Schwarz (2009). "Early-life programming of later-life brain and behavior: a
critical role for the immune system." Front Behav Neurosci 3: 14. Bode, L., R. Ferszt, et al. (1993). "Borna disease virus infection and affective disorders in man."
Arch Virol Suppl 7: 159-167. Boksa, P. (2010). "Effects of prenatal infection on brain development and behavior: A review of
findings from animal models." Brain, Behavior, and Immunity 24(6): 881-897. Ekdahl, C. T., J.-H. Claasen, et al. (2003). "Inflammation is detrimental for neurogenesis in adult
brain." Proceedings of the National Academy of Sciences 100(23): 13632-13637. Fatemi, S. H., J. Earle, et al. (2002). "Prenatal viral infection leads to pyramidal cell atrophy and
macrocephaly in adulthood: implications for genesis of autism and schizophrenia." Cell Mol Neurobiol 22(1): 25-33.
Fatemi, S. H., T. J. Reutiman, et al. (2008). "Maternal infection leads to abnormal gene
regulation and brain atrophy in mouse offspring: implications for genesis of neurodevelopmental disorders." Schizophr Res 99(1-3): 56-70.
Kessler, R. C., W. T. Chiu, et al. (2005). "Prevalence, severity, and comorbidity of 12-month
DSM-IV disorders in the National Comorbidity Survey Replication." Arch Gen Psychiatry 62(6): 617-627.
Nelson, M. D., A. J. Saykin, et al. (1998). "Hippocampal Volume Reduction in Schizophrenia as
Assessed by Magnetic Resonance Imaging: A Meta-analytic Study." Arch Gen Psychiatry 55(5): 433-440.
Rantakallio, P., P. Jones, et al. (1997). "Association between central nervous system infections
during childhood and adult onset schizophrenia and other psychoses: a 28-year follow-up." Int J Epidemiol 26(4): 837-843.
Rosoklija, G., G. Toomayan, et al. (2000). "Structural Abnormalities of Subicular Dendrites in
Subjects With Schizophrenia and Mood Disorders: Preliminary Findings." Arch Gen Psychiatry 57(4): 349-356.
Shah, D. K., L. W. Doyle, et al. (2008). "Adverse neurodevelopment in preterm infants with
postnatal sepsis or necrotizing enterocolitis is mediated by white matter abnormalities on magnetic resonance imaging at term." J Pediatr 153(2): 170-175, 175 e171.
4
Stoll, B. J., N. I. Hansen, et al. (2004). "Neurodevelopmental and growth impairment among
extremely low-birth-weight infants with neonatal infection." JAMA 292(19): 2357-2365. van Berckel, B. N., M. G. Bossong, et al. (2008). "Microglia Activation in Recent-Onset
Schizophrenia: A Quantitative (R)-[11C]PK11195 Positron Emission Tomography Study." Biological Psychiatry 64(9): 820-822.
5
CHAPTER 2
LITERATURE REVIEW
2.1 Neurodevelopment of the Domestic Pig
The process of neurodevelopment begins very early in embryogenesis and is relatively
conserved across mammalian species. Much is known about pig fetal development as it is a
standard for mammalian embryology (Patten 1931). After fertilization, embryo implantation
occurs in the uterine wall of the sow by the 14th day of pregnancy (Book and Bustad 1974). The
pig has a flat embryonic disk and the neural plate is formed on the dorsal surface of the embryo
from ectodermal tissue at 14 days. Soon after formation, the neural plate invaginates and
becomes folded forming the neural tube. Neural tube closure is similar to humans as it is
initiated at multiple points along the anteroposterior axis (van Straaten, Peeters et al. 2000).
After neural tube closure, roughly 17 days post-conception, the anterior portion dilates and
enlarges to form the prosencephalon, mesencephalon, and rhombencephalon. By 20-22 days,
the brain has divided into five regions including the telencephalon, diencephalon,
mesencephalon, metencephalon, and myelencephalon. At this time, cortical development
begins with the formation of the preplate and subsequent increases in neurogenesis and
neuronal migration. One critical protein for cortical formation and organization is Reelin. Reelin
expression is highest in the Cajal-Retzius cells within the marginal zone. The highest levels
occur in the pig at E60 and correspond to peak neurogenesis (Nielsen, Sondergaard et al.
2010). Expression patterns of Reelin in the pig are very similar to humans.
Much of the brain development characterization in late gestation in the pig has been
conducted in larger regional areas with a few studies on cellular development. Early studies by
Dickerson and Dobbing, and confirmed by Pond et al., characterized brain weight and
cholesterol content from the 52nd day of gestation through 3 years (Dickerson and Dobbing
6
1967; Pond, Boleman et al. 2000). Both humans and pigs have a perinatal growth spurt with
the brain being 27% and 25% of adult weight at birth respectively (Dobbing and Sands 1979).
This is dissimilar to other species such as the rhesus monkey and rodents which have prenatal
and postnatal growth spurts respectively. Passingham compared regional brain development
between many mammals and found that ungulates, including the pig, and primates had very
similar brain growth during gestation, but ungulates build their body at a higher growth rate
(Passingham 1985). Comparison of subareas of the brain between the pig and rhesus monkey
revealed similar proportions in all but two areas. The pig has a larger paleocortex, including
amygdala, and the rhesus monkey has a proportionally higher amount of neocortex. Magnetic
resonance imaging studies have characterized the normal postnatal brain development of the
pig from 2- to 24-weeks of age (Conrad, Dilger et al. 2012). During this time, the whole brain
volume increases 120-130% with a sexual dimorphism in regional growth similar to humans.
Stereological studies conducted by Jelsing et al. found significant postnatal gliogenesis in both
domestic and Göttingen pigs, but the Göttingen minipig also had significant neurogenesis
(Jelsing, Nielsen et al. 2006). The authors concluded that the domestic pig may be a better
model for developmental hypotheses than the Göttingen minipig. An extensive list of
neuroanatomical studies of the porcine brain, including functional cortical mapping, has been
previously published (Lind, Moustgaard et al. 2007).
In addition to neuroanatomical characterization of brain development in the pig,
immunohistochemistry has been used to describe neurotransmitter system architecture and
function. The two most well characterized systems in the pig are the serotonergic and
dopaminergic systems. Tryptophan hydroxylase serotonergic (5-HT) neuron distribution in the
medulla is very similar to the human infant, with a few differences such as additional clusters on
the ventral surface in the arcuate nucleus (Niblock, Luce et al. 2005). Age-related differences
are also seen in both the pig and human, where in humans the distribution of 5-HT cells change
while receptor binding changes in the pig. Distributions of dopaminergic cells in the pig have
7
been found to be very similar between human and non-human primates (Lind, Moustgaard et al.
2007). Like humans, the highest amount of dopamine was found in the substantia nigra and
ventral tegmental area with a gradient distribution of dopamine receptor concentration (Larsen,
Bjarkam et al. 2004; Rosa-Neto, Doudet et al. 2004). Less is known about adrenergic and
noradrenergic systems, but it has been shown that highest levels occur in the diencephalon and
brainstem, similar to other mammals (Agarwal, Chandna et al. 1993). Interestingly, the pig has
a 10-fold higher expression of monoamine oxidase B (MAO-B) than MAO-A, enzymes that
catabolize catecholamines (Anderson, Lupo et al. 2005). This distribution of MAO isoforms is
seen in humans, but not rodents.
There are several differences in pregnancy and embryogenesis between humans and
pig with the majority attributed to different timescales of development. Embryonic implantation
occurs earlier in humans than pigs, by the 6th day after ovulation (Book and Bustad 1974). The
placental layers differ where humans have a discoid hemochorial placenta allowing direct
contact between the fetal chorion and maternal blood (Book and Bustad 1974). Pigs have a
diffuse epitheliochorial placenta. In addition, humans usually have one child while pigs are
multiparous. The gestational length for humans is longer, 40-weeks compared to 114 days. The
time course for intrauterine development is slightly different in humans and pigs, with the human
having a more advanced development when corrected for percent of gestation (Book and
Bustad 1974). But within the developmental stages, the fetus is very similar. This difference
allows the human fetus to have more time in the third fetal stage.
2.2 Neuroimaging in the Piglet
Magnetic resonance imaging (MRI) is a safe, non-ionizing radiation modality that is ideal for in
vivo neuroimaging both clinically and for research. Different MRI sequences allow imaging and
quantification of brain region anatomy, function, and neurochemical content and are very useful
8
for serial studies including tracking neurodevelopment. The use of MRI in the pig has lagged
behind research using humans, non-human primates, and rodents, but these techniques are
easily transferable to a pig model. Due to the size of the piglet, clinical strength magnets with
head coils designed for humans are able to be used with piglets without modification. Rodents,
however, require specially designed magnets at higher field strength due to the size of the
animal. Thus, the piglet represents an ideal animal model for using clinical MRI scanners.
One of the most common uses for MRI is to image the structure of the brain. A
combination of both T1- and T2-weighted imaging sequences are used clinically to examine
brain structure and to identify abnormalities or pathologies. These sequences are also very
useful for research. One of the first studies using MRI in pigs was conducted in 1993 where the
authors acquired structural images from pigs placed in stereotaxic conditions and then
compared the images to postmortem histological sections (Marcilloux, Felix et al. 1993). This
work was further expanded upon and refined in 2000 in order to correlate histological data onto
the MR images (Sørensen, Bjarkam et al. 2000). This was an important step as prior to 2000,
only a histologically based atlas of the domestic pig was available, and this paved the way for
future work in creating digital MR-based structural atlases of the brain (Felix, Leger et al. 1999).
One of the first MR-based atlases was constructed in the Göttingen minipig using 22 subjects
and labeling 34 regions of interest (Watanabe, Andersen et al. 2001). This study opened the
door to cross-modality analysis such as regional quantification of radiotracer uptake in positron
emission topography (PET) studies (Rosa-Neto, Gjedde et al. 2004). Unfortunately, this atlas is
not appropriate for analysis of the domestic pig brain due to the size and morphology
differences between the Göttingen minipig brain and the domestic pig. A high resolution MR-
based atlas for the adult domestic pig was constructed in 2009 by Saikali et al. and is a very
good resource for studies using adult pigs (Saikali, Meurice et al. 2010). This atlas is not
appropriate for studies in piglets though due once again to size differences between the
neonatal and adult brain. Previous studies using MRI in piglets used manual segmentation
9
methods for brain region volume estimation (Conrad, Dilger et al. 2012). Using these
techniques, a longitudinal study was conducted characterizing the normal brain growth and
development of the domestic pig from 2- to 24-weeks of age showing many similarities and a
few differences in brain growth compared to humans (Conrad, Dilger et al. 2012). Additionally,
brain growth curves for the domestic pig have also been made using a cross-section study
design (Winter, Dorner et al. 2011). Recently, an averaged brain and atlas for the domestic pig
was constructed that will allow for more sophisticated analysis techniques, such as voxel based
morphometry, to be conducted in the piglet (http://pigmri.illinois.edu). The use of these MRI
techniques will allow for hypothesis driven research in the pig such as quantification of brain
structure changes due to environmental, nutritional, and inflammatory insults.
In addition to T1- and T2-weighted structural imaging, diffusion tensor imaging (DTI) and
diffusion-weighted imaging (DWI) provides information about brain microstructure and white
matter tract structure. Diffusion tensor imaging uses diffusion properties of water molecules to
produce contrast by quantifying the magnitude, degree of anisotropy, and orientation of diffusion
anisotropy (Alexander, Lee et al. 2007). Diffusion anisotropy and the principle diffusion
directions are also used for white matter tractography (Alexander, Lee et al. 2007). These
measures can track microstructure changes the brain and are very useful for developmental and
aging studies, as well as clinically looking at brain pathologies. The three parameters that are
generated from diffusion tensor imaging are the three diffusion tensor eigenvalues, mean
diffusivity (MD), and fractional anisotropy (FA), which collectively describes the direction of
diffusion and magnitude of anisotropy. One of the first papers to use DWI in the piglet
characterized apparent diffusion coefficients (ADC), now reported as MD, in both grey and white
matter after perinatal-asphyxia (Thornton, Ordidge et al. 1997). The changes in MD and FA
values during normal development have also been characterized in domestic pigs and found
similar trends in changes in FA over time but dissimilar trends in MD compared to humans
(Winter, Dorner et al. 2011). The differences in MD levels may be due to different timing of
10
water content changes in the pig. DTI and DWI provide robust information about brain
microstructure and white matter development/integrity and are well suited for use in a pig model.
MRI also allows for in vivo quantification of brain metabolites. 1H magnetic resonance
spectroscopy (MRS) is the most commonly used spectroscopy technique and analyzes the
signal from hydrogen protons that are bound to different chemicals. Detectible metabolites
include N-acetyl-aspartate, creatine, choline, myo-inositol, as well as neurotransmitters such as
glutamate and glutamine to name a few. Spatial resolution is reduced in MRS compared to
MRI, thus spectra that are obtained are for larger, more encompassing areas (Blüml 2013).
There are two methods used for MRS, single voxel MRS and chemical shift imaging. Single
voxel MRS quantifies metabolites in one region of the brain and is able to quantify more
metabolites, including lower concentration metabolites, due to a better signal to noise ratio
(S/N). Chemical shift imaging is able to acquire spectra from multiple voxels, but at the expense
of S/N, meaning less detection of low concentration metabolites. MRS has been used in pigs in
both brain trauma and hypoxia-ischemia research to detect changes to metabolites in the brain
(Smith, Cecil et al. 1998; Munkeby, De Lange et al. 2008).
A lesser used MRI modality in pigs is functional magnetic resonance imaging (fMRI).
fMRI detects changes in brain activity by analyzing differences in the blood-oxygen-level
dependent (BOLD) response , which detects increases to blood flow to a region. Studies have
been conducted using conscious pigs in order to look at regional activation in response to pain
and visual stimulation (Fang, Lorke et al. 2005; Fang, Li et al. 2006). Even though it has had
limited use in pigs, fMRI is a valuable tool for translational imaging studies.
2.3 The Piglet as a Model for Human Neurodevelopment
Workman et al. have constructed a very good resource allowing for comparison of 271
neurodevelopmental events to be translated across species (Workman, Charvet et al. 2013).
11
Based on developmental allometry, highly predictive models for timing of neurodevelopmental
events were constructed and validated in multiple species. Additionally, their website
(www.translatingtime.net) includes instructions for estimating the event timing in new species
which we have done for the pig. For this calculation, we used a gestational length of 114 days
and an adult brain weight of 180 grams. Using these parameters, the species constant for the
domestic pig equaled 2.984 and the slope 2.809. Using the modeling equation and event
scores, estimations of developmental equivalencies can be computed between species. Using
the MRI-based whole brain volume estimates for the pig, we compared this to the translating
time estimates for whole brain weight for the pig and human. Linear regression of the
translating time data estimated that one pig week is roughly equal to 3 human weeks. We
estimate through our MRI data, that one pig week is equal to one human month. The authors of
translating time indicate that the model is based only on data up to 3 years old in human and
thus only applicable to prenatal and early-postnatal development. Similarly, we found their
model to fit our data well around the perinatal period with larger deviations as the animal grew
towards adulthood.
Additionally, Niblock et al. estimated the comparable age of the piglet to humans based
on serotonergic system development (Niblock, Luce et al. 2005). They suggest that a 4, 12, 30,
and 60 day old pig is similar to a 1, 4, 6, and 12 month old human. From these data, rough
estimations can be made for equivalent ages during the early neonatal period between pigs and
humans, but more work is needed to validate these estimations.
2.4 Piglet Models in Neonatal Research
Infection Models
12
The pig has striking similarities in immune system anatomy, function, and gene
expression compared to humans (Freeman, Ivens et al. 2012; Meurens, Summerfield et al.
2012). One study examined immune system similarities between pigs, humans, and rodents
and found that the pig was more similar than rodents to humans in 80% of analyzed parameters
(Dawson 2011). As one example in the innate immune response, macrophages stimulated with
lipopolysaccharide (LPS) produce nitric oxide in rodents, but this response is not seen in both
human and pig macrophages (Pampusch, Bennaars et al. 1998). Because these similarities,
the pig presents as important translational animal model that may be more predictive of
therapeutic interventions than other models.
The pig has been used extensively for studying pathogens which cause systemic,
respiratory, and gastrointestinal tract infections (Meurens, Summerfield et al. 2012). Many of
the same pathogens that impact human health are highly prevalent in pigs and are important
both from a biomedical research and a pork production standpoint. Staphylococcus aureus,
including methicillin-resistant S. aureus, is becoming a worldwide clinical problem and many
studies have been conducted in pigs examining wound infection, osteomyelitis, and pneumonia
(Svedman, Ljungh et al. 1989; Luna, Sibila et al. 2009; Jensen, Nielsen et al. 2010). Other
human pathogens, such as Bordetella pertussis, are able to infect piglets and present as a very
good model for direct translational research (Elahi, Buchanan et al. 2006). Gnotobiotic piglets
are the only animal model that is susceptible to human rotovirus, and this model has been used
to research pathogenicity and vaccination approaches (Saif, Ward et al. 1996; Meurens,
Summerfield et al. 2012).
Hypoxia-ischemia Models
Due the body weight and size of the piglet, it has a rich history of being used as a model
for human neonatal hypoxia ischemia (HI) (Roohey, Raju et al. 1997). Much of the model
13
development was conduct in the 1980‘s including characterizing changes to cerebral blood flow
and metabolism. This research, along with fetal sheep studies, described the basic physiology
of the response to acute asphyxia including hyper/hypocapnia, hyper/hypoglycemia, and
hypovolemic hypertension (Raju 1992).
There are multiple approaches that can be taken to produce hypoxia ischemia
encephalopathy (HIE) that model different clinical situations. The first is a whole body approach
by reducing the inspired air to hypoxic conditions (i.e. FiO2 = 0.08) (Haaland, Loberg et al. 1997;
Bjorkman, Foster et al. 2006). This produces HI in both the brain as well as the peripheral
organs. A second approach uses the whole body HI and then includes an asphyxia component
resulting in cardiac arrest and resuscitation (Agnew, Koehler et al. 2003). A review by Alonso-
Spilsbury et al describes in detail pathological changes seen in asphyxia in both a biomedical
and pork production view (Alonso-Spilsbury, Mota-Rojas et al. 2005). A third approach is to
focus the HI insult just on the brain, excluding peripheral impact. For this, a bilateral carotid
ligation technique can be used (LeBlanc, Vig et al. 1991). The physiologic changes and brain
damage seen in the piglets are very similar to the human neonate, thus making the piglet an
ideal model for therapeutic development.
Magnetic resonance imaging allows for the study of brain injury severity in neonates
(Barkovich, Miller et al. 2006). Similar imaging protocols are able to be used in piglet models of
HIE. HIE causes energy and metabolism disruptions and these can be quantified using
magnetic resonance spectroscopy (Thornton, Ordidge et al. 1997; Vial, Serriere et al. 2004). In
addition, diffusion-weighted imaging allows for the quantification of changes in water diffusion
following HIE insult (Thornton, Ordidge et al. 1997; Cheng, Liu et al. 2005; Munkeby, De Lange
et al. 2008). MR angiography can also be used to track perfusion changes during the
experimental protocol (Munkeby, Lyng et al. 2004). The direct translation of these imaging
methods strengthens the piglet as an animal model for HIE.
14
An example of how research conducted in the piglet has directly translated to
improvements in neonatal care is exemplified by hypothermia treatment for neonatal HIE. HIE,
which previously had no treatments, causes high neonatal mortality rates and long-term
neurological complications (Higgins, Raju et al. 2011). Research conducted by Thoresen and
others have used a piglet model of HIE to determine timing of administration and improvements
to survival and brain damage using hypothermia therapy (Thoresen, Haaland et al. 1996;
Haaland, Loberg et al. 1997). Using these time points as guides, clinical studies were
conducted which showed improvements to survival and neurodevelopmental disability
(Shankaran, Pappas et al. 2012; Tagin, Woolcott et al. 2012; Thoresen, Tooley et al. 2013).
Based on these studies, the standard clinical practice is to now offer hypothermia therapy for
patients with HIE whom meet the inclusion criteria (Zanelli, Naylor et al. 2008; Perlman, Wyllie
et al. 2010).
2.5 Microglia and the Developing Brain
Microglial cells are the resident macrophages of the brain and were originally identified
by Nissl in 1899 who considered them to be reactive neuroglia with the capacity for migration
and phagocytosis (Barron 1995). Rio-Hortega advanced the knowledge of microglia by
suggesting that they were of mesenchymal origin and were non-neuronal cells (Barron 1995). It
is now known that approximately 10-12% of the cells in the brain are microglia and the regional
distribution varies with the highest density in the hypothalamus, basal ganglia, substantia nigra,
and hippocampus (Lawson, Perry et al. 1990; Mittelbronn, Dietz et al. 2001). Microglia share
many qualities with peripheral macrophage including expression of the markers CD11b, CD14,
and F4/80, but they are not of bone marrow origin. Studies in rodents have shown that
microglia progenitor cells originate in the extra-embryonic yolk sac and migrate into the CNS via
blood circulation between E8.5 and E9.5 (Ginhoux, Greter et al. 2010). The lifespan of
15
microglia is unknown but studies have shown that the microglia population can expand under
pathological conditions (Saijo and Glass 2011). There is a sexual dimorphism in microglia
number and morphology in the developing brain (Schwarz, Sholar et al. 2012). Males have
more microglia early in development and females have an activated phenotype later into
development. This dimorphism may be a contributing factor in the sex bias of many
neurodevelopmental disorders.
Microglia have been shown to have a multifaceted role in the CNS including
maintenance of normal tissue homeostasis, response to inflammation and pathogens, and
normal brain development. Microglia have two activation states which can be characterized by
cell morphology. During the “resting” or quiescent state, microglia exhibit a small soma with
long and thin processes (Hanisch and Kettenmann 2007). Microglia act as surveying cells
sampling the CNS microenvironment and display high levels of motility (Nimmerjahn, Kirchhoff
et al. 2005). In early development and in response to inflammatory stimuli, microglia enter an
active phase, retract their processes, resulting in an amoeboid cell morphology, and exhibit
phagocytic actions (Ransohoff and Perry 2009). Pro- and anti-inflammatory signals modulate
the microglia response. Microglia express a variety of cytokine and chemokine receptors which
sense the inflammatory environment and allow the cells to mount an appropriate response
(Saijo and Glass 2011). Similar to peripheral macrophages, microglia are capable of displaying
two different phenotypes, M1 (classical) and M2 (alternative) activation. The M1 phenotype is
displayed in response to pro-inflammatory conditions including activation of Toll-like receptors
(TLRs) and interferon-γ (Michelucci, Heurtaux et al. 2009). In response, microglia up-regulate
major histocompatibility markers (MHC class I and II) and produce proinflammatory cytokines
and free radicals (Perry and Gordon 1988; Gordon 2003). The alternative (M2) phenotype is
stimulated with IL-4 and IL-13 and is marked by arginase 1 expression (Gordon 2003). The M2
phenotype is involved with tissue remodeling and wound repair (Song, Ouyang et al. 2000). IL-
16
10, an anti-inflammatory cytokine, strongly enhances the phagocytic activity of microglia which
may contribute to wound repair (Michelucci, Heurtaux et al. 2009). Although microglia are
essential for host defense and neuroprotection, over activation and dysregulation during
development may be a “vulnerability” factor for later-life pathology (Bilbo and Schwarz 2009).
In the adult brain, the majority of microglia are in the resting state. During development
however, microglia are reactive and amoeboid, and have a crucial role in normal brain
development (Cuadros and Navascues 1998). Microglia participate in many developmental
processes including synaptic remodeling, clearance of dying neurons, and contribute to
neuronal proliferation and differentiation (Harry 2013). Microglia participate in synaptic
remodeling through release of neurotrophic factors as well as physically modifying the
synapses. Microglia produce and release multiple neurotrophic factors including neuronal
growth factor (NGF), brain derived neurotrophic factor (BDNF), neurotrophin-3 (NT3), and glial
derived neurotrophic factor (GNDF), which induce and modify synaptic plasticity (Kim and de
Vellis 2005; Harry 2013). Microglial processes are found in synaptic clefts where they modulate
synaptic strengthening and degradation and are important for shaping the visual cortex during
early life (Rochefort, Quenech'du et al. 2002; Tremblay, Lowery et al. 2010). Additionally,
microglia participate in synaptic pruning by engulfing and removing dendritic spines (Paolicelli,
Bolasco et al. 2011). Microglia appear in the brain parenchyma at a time when programed cell
death is beginning and apoptotic cells start to appear (Wakselman, Bechade et al. 2008; Rigato,
Buckinx et al. 2011). CD11b and DAP12 expressed by microglia control the production of
superoxide molecules which promote neuronal cells death (Wakselman, Bechade et al. 2008).
Loss or blockade of either CD11b or DAP12 leads to decreases in developmental apoptosis.
Microglia also contribute to neuronal proliferation through guidance of neuronal progenitor cells
and influence differentiation (Aarum, Sandberg et al. 2003). They have been shown to promote
17
differentiation of cholinergic neurons in the basal forebrain and drive a switch from neurogenesis
to astrocyte formation (Jonakait, Wen et al. 2000; Antony, Paquin et al. 2011).
2.6 Immune to Brain Communication
Historically, the brain has been considered an immune-privileged organ, but this is now
known to be untrue (Dantzer, O'Connor et al. 2008). Within the choroid plexus and meninges,
there are multiple immune cells including macrophages and dendritic cells (Dantzer, O'Connor
et al. 2008). The brain parenchyma has its own distinct subset of macrophages, microglia cells,
which are able to detect and respond to inflammatory signals. In addition to microglia cells,
neurons , astrocytes, and oligodendrocytes have cytokine receptors (Sawada, Itoh et al. 1993).
These receptors allow for the sensing of inflammation and contribute to the bi-directional
communication between the immune system and central nervous system.
This bi-directional communication allows for a “sixth sense” which is important for
modifying physiology and behavior in response to injury or infection. In the periphery, the innate
immune system has a variety of mechanism for detecting potential pathogens called pattern-
recognition receptors (PRRs) which respond to specific moieties or pathogen-associated
molecular patterns (PAMPs) (Dantzer 2009). Upon activation of the PRRs, multiple molecular
pathways are initiated to combat the pathogen. The majority of these pathways ultimately lead
to the up-regulation nuclear factor kappa-light-chain-enhancer of activated B cells (NFκB)
(Kumar, Kawai et al. 2011). NFκB activation leads to increased synthesis and release of the
pro-inflammatory cytokines interleukin (IL)-1β, IL-6, and tumor necrosis factor-α (TNF-α)
(Pugazhenthi, Zhang et al. 2013).
Cytokines that are produced in the periphery are able to act on the brain through both
neural and humoral pathways to modify behaviors classically termed “sickness symptoms”
18
(Dantzer, Konsman et al. 2000; Dantzer and Kelley 2007). Sickness symptom behaviors
include anorexia, listlessness, fatigue and malaise, sleep disturbances, and social withdrawal
(Kelley, Bluthe et al. 2003). These behavior modifications are thought to provide protection and
reduce energy expenditures allowing the body to focus on fighting the infection. In addition to
behavior modifications, physiological changes occur to aid in the clearance of the infection
including fever and hypothalamic-pituitary-adrenal axis (HPA) activation (Johnson 2002). Fever
induction is caused by pro-inflammatory cytokines activating cells within the preoptic area of the
hypothalamus and aids in the clearance of pathogens, especially bacteria (Conti, Tabarean et
al. 2004).
In order to induce a behavioral response, peripheral cytokines must pass their signal to
the brain. This is accomplished via neural and humoral pathways. In the periphery, the afferent
vagus nerve has cytokine receptors and can transmit information about the inflammatory state
to the nucleus tractus solitarius and secondary projection areas of parabrachial nucleus,
hypothalamic paraventriculum, and supraoptic nuclei (Wan, Janz et al. 1993). Here, the neural
signals activate cytokine expression centrally causing sickness behavior (Goehler, Gaykema et
al. 1998). Subdiaphragmatic vagotomy attenuates some behavioral responses but leaves the
fever response functional indicating multiple messaging pathways for immune-to-brain
communication (Kapcala, He et al. 1996; Romanovsky, Simons et al. 1997). Cytokines can also
access the brain by humoral pathways through active transport mechanisms and diffusion at
circumventricular organs where there is a lack of a blood brain barrier (Banks, Farr et al. 2002).
Endothelial cells that compose the blood brain barrier are capable of detecting pro-inflammatory
cytokines and propagate the signal by release of cytokines and prostaglandins into the brain
(Van Dam, De Vries et al. 1996). Once these pro-inflammatory cytokines reach CNS, they can
activate microglia cells which will amplify and transmit the signal.
19
2.7 Pathogenesis of Porcine Reproductive and Respiratory Syndrome Virus
Porcine reproductive and respiratory syndrome virus (PRRS) was first described as
“mystery swine disease” and “blue ear disease” when it emerged in North America and Europe
in the 1980’s (Done, Paton et al. 1996). The virus was formally named and characterized in the
early 1990’s as belonging to the Arteriviridae family of enveloped viruses (Dokland 2010).
PRRS infection is characterized by marked interstitial pneumonia and reproductive losses and
increases the risk for secondary infections (Rossow, Shivers et al. 1999). The economic impact
of PRRS on the pork production industry was estimated to be $664 million annually as of 2013
(Holtkamp, Kliebenstein et al. 2013).
Due to the impact of this disease, much research has been conducted into the structure
and pathogenicity of the virus. The virus contains a positive-sense (+) RNA genome with a
length of 15.1 – 15.5 kb (Dokland 2010). Seven open reading frames (ORF) allow for the
coding of 14 non-structural proteins, including proteases, a RNA-dependent RNA polymerase,
and a helicase, and multiple structural proteins (Amonsin, Kedkovid et al. 2009). PRRS has a
very narrow cell tropism, first infecting alveolar macrophages in the lungs and then spreading to
macrophages in other areas (Van Breedam, Delputte et al. 2010). The PRRS virion first
attaches to the porcine macrophage by binding to heparin sulfate GAGs on the surface of the
macrophage. Next, the M/GP5 glycoprotein on the virion envelope binds to a sialoadhesin
receptor and is internalized by clathrin-mediated endocytosis (Van Breedam, Van Gorp et al.
2010). Once internalized, porcine CD163 interacts with GP2 and GP4 on the virion envelope
allowing for the release of the viral genome (Calvert, Slade et al. 2007). When the viral genome
is within the cell, it replicates, assembles new virion particles, and then exits through a budding
process (Zhang, Xue et al. 2010).
20
The innate immune system has multiple ways to detect and fight PRRS viral infection.
The Toll-like receptors (TLRs) are proteins of the innate immune system that detect distinct
pathogen-associated molecular patterns (PAMPs). In response to TLR activation, specific
signaling pathways are activated to induce expression of pro- and anti-inflammatory cytokines to
combat the infection. TLR3, 7, and 8 are located on the endosome membrane and respond to
viral infection. Intracellular double stranded RNA (dsRNA) activates the TLR3 receptor. Upon
binding of dsRNA, TLR3 activates interferon regulatory factor 3 (IRF-3) which induces type I
interferon and NF-κB. Single stranded RNA (ssRNA) bind to TLR7 and 8 inducing MyD88 and
IRF-7. Similarly to IRF-3, IRF-7 activates type I interferon and NF-κB. The PRRS virus is able
to activate TLR3, 7, and 8 as both ssRNA and dsRNA are produced during the viral genome
replication (Zhang, Liu et al. 2013).
2.8 PRRSV and Neuroinflammation
The majority of PRRS research has looked at the peripheral aspects of the infection with
the goal of developing a vaccine or improving pork production. Upregulation of pro-
inflammatory cytokines in the periphery, including IL-6 and IL-1β, is a physiological response by
the innate immune system to fight the infection. Through immune-to-brain communication, the
brain can sense the inflammatory state of the periphery and reacts with changes to behavior
and gene expression in the brain (Dantzer, O'Connor et al. 2008). In addition to
neuroinflammation induction through the periphery, some strains of PRRS may have
neurovirulence. PRRS has been shown to co-localize with MAC-387 positive cells (macrophage
marker) in brain tissue (Rossow, Shivers et al. 1999). It is unknown if these are microglia cells
or invading monocytes/macrophages from the periphery.
Evidence of peripheral and central immune activation in neonatal and young piglets has
been well characterized (Miguel, Chen et al. 2010; Elmore, Burton et al. 2014). Experimentally
21
inoculating piglets with a highly-virulent strain of PRRS elicits a strong peripheral cytokine
response with increased expression of TNF-α, IL-1β, IL-10, IL-6 and IFN-γ in the blood (Miguel,
Chen et al. 2010; Zhang, Liu et al. 2013). Pro-inflammatory cytokines and TLR receptors are
also up-regulated in multiple regions of the brain. The source of these pro-inflammatory
cytokines may be from activated microglia or reactive astrocytes (Lieberman, Pitha et al. 1989;
Dantzer, O'Connor et al. 2008). Increases in IL-10 and reductions in CD200, nerve growth
factor (NGF), and myelin basic protein (MBP) were also evident in the brain of PRRS-infected
piglets (Elmore, Burton et al. 2014).
We have previously shown that PRRS infection in piglets leads to drastic microglia
activation in the hippocampus with 82% and 43% of isolated microglia being MHC-II positive 13
and 20 days post inoculation (Elmore, Burton et al. 2014). The presence of MHC-II shows that
the microglia are activated, but it does not differentiate between the classical M1 (inflammatory)
versus alternative M2 (anti-inflammatory) phenotype (Kigerl, Gensel et al. 2009; Olah, Biber et
al. 2011). Further characterization of the phenotype of these microglia will help the determine
timing of the inflammatory state and whether there is a M1 to M2 switch during disease
progression.
The increase in pro-inflammatory cytokines has the potential to impact normal brain
development processes and produce deficits in learning and memory. The hippocampus is very
important for learning and memory and is susceptible to neuroinflammation (Yirmiya and
Goshen 2011). Using a hippocampal dependent spatial t-maze task, piglets with active PRRS
infection show deficits in learning and memory (Elmore, Burton et al. 2014). PRRS infection
increased the number of days that the piglets needed to acquire the task. The PRRS piglets
also exhibited increased latency to choice and distance moved in the maze showing that they
completed the task, but not to the performance level of the controls. There are multiple
mechanisms that can explain the learning and memory deficits including changes to synaptic
22
plasticity, neurogenesis, and neuron morphology. The role of inflammation in modulating
neurogenesis and neuron morphology are discussed below.
2.9 Inflammation and Neurogenesis
The timelines of development of the human and rodent brain has been well
characterized. Early in neurodevelopment, there is a high level of neuronal proliferation and
migration that lasts until the early postnatal period (Rice and Barone 2000). The progenitor cells
that form these new neurons undergo a switch during mid-gestation and start producing glial
cells, including astrocytes and oligodendrocytes. The formation of these cells persists into the
postnatal period and even into early adulthood (Rice and Barone 2000). By birth, the majority of
neurogenesis has been completed, but there are two areas of the brain which neurogenesis
continues throughout life. The subependyma of the lateral ventricles produce new neurons that
migrate in the rostral stream into the olfactory bulb and the subgranular zone of the dentate
gyrus (DG) produces dentate granule cell neurons (Ekdahl 2012). Much research has been
conducted into how inflammation can impact neurogenesis, cell fate, and behavioral
consequences in adults. A few studies have looked at prenatal inflammation’s effects on
neurogenesis, but there is a void of studies looking at the early postnatal period.
Neural progenitor cells in the DG have a limited-self renewing capacity and produce
throughout life (Kempermann, Gast et al. 2003). Of the newly born cells, the majority will die but
a few will migrate into the DG granule cell layer and mature into functional neurons (Biebl,
Cooper et al. 2000; Kempermann, Gast et al. 2003). The exact role of these new neurons is
unknown, but they may be crucial for hippocampal-dependent forms of spatial memory
(Nakashiba, Cushman et al. 2012). Inflammation has been shown to alter adult hippocampal
neurogenesis in a variety of ways including proliferation, survival, differentiation, and
incorporation into new networks (Belarbi and Rosi 2013; Kohman and Rhodes 2013). The first
23
studies to show a link between inflammation and changes in neurogenesis were conducted in
the early 2000s by Monje et al. and Ekdahl et al. Both studies found that LPS administration,
which leads to classical microglia activation, resulted in decreased new neuron survival but no
changes in proliferation (Ekdahl, Claasen et al. 2003; Monje, Toda et al. 2003). Minocycline,
which inhibits microglia activation, was able to block this decrease in neurogenesis (Ekdahl,
Claasen et al. 2003). Classical microglia activation leads to an upregulation of pro-inflammatory
cytokines including TNF-α, IL-1β, and IL-6 (Michelucci, Heurtaux et al. 2009). In vitro and in
vivo studies have shown that these pro-inflammatory cytokines are able to neural precursor
generation, differentiation, and survival (Vallieres, Campbell et al. 2002; Cacci, Ajmone-Cat et
al. 2008; Wu, Hein et al. 2012; Wu, Montgomery et al. 2013). Some studies have also shown
that pro-inflammatory cytokines can reduce the proliferation rate (Fujioka and Akema 2010).
Alternative activation of microglia produce the opposite effect. IL-10, IL-4, TGF-β, and IGF-1
promote production of neurons (Aberg, Aberg et al. 2000; Battista, Ferrari et al. 2006; Kiyota,
Okuyama et al. 2010; Kiyota, Ingraham et al. 2012). In addition to affecting proliferation and
survival, cytokines also drive changes in cell fate. Pro-inflammatory cytokines causes a switch
from neurogenesis to gliogenesis and increase astrocyte formation (Green, Treacy et al. 2012;
Stolp 2013). Inflammation can also impact the electrophysiological properties of newly
incorporated neurons (Jakubs, Bonde et al. 2008).
Although the majority of studies looking at the impact of neuroinflammation have been
conducted in adults, a few have looked at the impact of prenatal and neonatal inflammation.
Jӓrlestedt et al. administered LPS to mice at P9 and characterized long term survival of neurons
and astrocytes (Jarlestedt, Naylor et al. 2013). They found that LPS injection reduced the
number of surviving neurons and astrocytes at P41, and this was only true for the dorsal
hippocampus. No long term behavioral differences were found. Prenatal immune activation
has been shown to both decrease and increase neurogenesis. Prenatal LPS injection leads to
decreased cell survival in the offspring, but this is dependent on timing of LPS administration
24
(Cui, Ashdown et al. 2009). Alternatively, maternal E. coli infection causes increased number of
proliferating cells in the offspring and increased expression of BDNF, TrkB, and p-Akt (Jiang,
Zhu et al. 2013). In addition, rats from E. coli infected dams displayed spatial learning and
memory deficits.
From these studies, the complexity of the impact of inflammation on neurogenesis can
be seen. The phenomena seen in adult studies may not be predictive of what happens during
early development. In addition, the timing and duration of inflammatory insult may differentially
impact neurogenesis during development.
2.10 Inflammation, Neuron Morphology, and Dendritic Spines
Due to the role of the hippocampus in learning and memory, the structure and function of
the cells within this region have been well characterized (Olton, Becker et al. 1979; von Bohlen
Und Halbach 2009). Much of the early work characterized the normal dendritic complexity of
the dentate granule cells and CA region pyramidal cells and how this complexity changes in
pathological states or in different environments (Green and Juraska 1985). For example,
environmental enrichment causes increased complexity within dentate granule cells for female
rats, but male rats were unaffected by environmental enrichment (Juraska, Fitch et al. 1985).
Additionally, there is a sexual dimorphism in the isolated environment, where males have more
dendritic material per neuron than females. The complexity of dentate granule cells also varies
with region where neurons located in the deep 2/3 of the dentate granule cell layer are less
complex than those in the superficial 1/3 (Green and Juraska 1985). Dendritic complexity is
very plastic and can be influenced by a number of factors, including stress and inflammation,
and changes can lead to deficits in learning and memory (McEwen 1999). Stress, via action of
adrenal steroids, can lead to atrophy of dendrites in the hippocampus (Woolley, Gould et al.
1990). Peripheral and central immune activation via LPS can also alter dendritic branching and
spine density (Milatovic, Zaja-Milatovic et al. 2003; Richwine, Parkin et al. 2008). Using a
25
mouse model of influenza infection, Jurgens et al. showed alterations in CA1 pyramidal neuron
morphology and differential effects on morphology in dentate granule cells based on layer
location (Jurgens, Amancherla et al. 2012). Much less is known about how early-life
inflammatory insult can affect neuron morphology. One study conducted in rats administered
LPS prenatally at E15 (Baharnoori, Brake et al. 2009). The authors found long-term disruptions
in the complexity of the dendritic arbor and variable changes in dendritic length. Changes to the
normal morphology may be clinically relevant as alterations in neuron morphology and spine
density have been found in diseases including schizophrenia and autism (Glantz and Lewis
2000; Kolluri, Sun et al. 2005; Pickett and London 2005)
In addition to dendritic tree complexity, dendritic spines are also sensitive to
environmental changes including inflammation (Bitzer-Quintero and Gonzalez-Burgos 2012).
Dendritic spines are small protrusions on the neuronal dendrite that are active or potential sites
for synapses. They play a role in synaptic plasticity and have different physiological properties
based on size and shape (von Bohlen Und Halbach 2009). Pro-inflammatory cytokines are
capable of altering the physiological properties of neurons and dendritic spine development
(Kondo, Kohsaka et al. 2011). For example, a single injection of LPS in young mice is sufficient
to destabilize spines, increase spine turnover rate, and reduce spine density after eight weeks
(Kondo, Kohsaka et al. 2011). Additionally, influenza infection is able to reduce the spine
density of inner dentate granules cells (Jurgens, Amancherla et al. 2012).
Changes to neuron morphology and spine density may have acute and chronic
implications. Alterations have the potential to impact synaptic plasticity and could be an
underlying mechanism for deficits in hippocampal-dependent learning and memory (Neves,
Cooke et al. 2008; Kasai, Fukuda et al. 2010). Disruptions during the neonatal period may be of
particular importance as this period has the highest rate of synaptogenesis and could lead to
lifelong consequences (Rice and Barone 2000).
26
2.11 Minocycline and Neuroinflammation
Neuroinflammation during the neonatal period can occur with many conditions including
hypoxic-ischemic encephalopathy and TORCH infections (Khandaker, Zimbron et al. 2012; Liu
and McCullough 2013). Microglia within the CNS have a central role in the inflammatory
response, and therapeutics designed to inhibit microglia may be beneficial. Minocycline has
been shown to reduce microglia activation and provide anti-inflammatory and anti-apoptotic
actions. Minocycline is a second-generation tetracycline antibiotic that is able to cross the blood
brain barrier. Clinically, it is routinely used in humans to treat urinary tract infections and acne.
The low doses used for antibiotic treatment is safe for long term use. Because of this,
minocycline has been the target of much research into its effectiveness as a treatment for
neuroinflammation.
Although a precise mechanism of action has not been found for minocycline in the CNS,
multiple actions may collectively provide neuroprotection. Minocycline has been shown to
inhibit the proliferation and activation of microglia cells including reductions in inflammatory
molecules including matrix metalloproteases, nitric oxide, and pro-inflammatory cytokines (He,
Appel et al. 2001; Stirling, Koochesfahani et al. 2005). Minocycline prevents p38 MAPK
signaling in microglia, which is crucial for microglia activation (Walton, DiRocco et al. 1998; Guo
and Bhat 2007). Additionally, minocycline provides anti-oxidant and anti-apoptotic actions.
Inhibition of caspase-1, -3, and cytochrome c release all provide anti-apoptotic effects (Chen,
Ona et al. 2000; Kim and Suh 2009). Minocycline also increases expression of Bcl-2 which
prevents cell death (Wang, Wei et al. 2004). Collectively, these actions may provide the
neuroprotective effects seen.
Because of these actions, minocycline has been researched extensively as a therapeutic
for neuroinflammatory disease in both humans and animal models. Animal models have shown
27
both protective and harmful effects in models of ischemia, multiple sclerosis, amyotrophic lateral
sclerosis (ALS), and Alzheimer’s as examples (Kim and Suh 2009). High minocycline doses
were found to be protective in a rat model of hypoxia ischemia, but exacerbated brain injury in a
mouse model showing that minocycline may have species specific responses (Tsuji, Wilson et
al. 2004; Fan, Pang et al. 2005). Issues with the proper dosage are problematic. The
neuroprotective dose in rodents is very high (45-90 mg/kg/day) compared to the clinically used
dose in humans (2-4 mg/kg) and this makes difficult in translating work done in rodent models to
human clinical trials (Buller, Carty et al. 2009). Nonetheless, many clinical trials have been
completed or are in progress using minocycline for neurodegenerative and psychiatric
conditions using low doses. Completed studies have shown mixed results. Minocycline caused
faster deterioration in ALS patients but improved negative symptoms in schizophrenia patients
(Gordon, Moore et al. 2007; Chaudhry, Hallak et al. 2012). Differences in the inflammatory
mechanism of these diseases may contribute to the differential responses to minocycline.
The use of minocycline in pediatric populations has been limited in humans due to side
effects. Tetracyclines act as ion chelators which bind calcium phosphate and can be
incorporated into developing teeth and bone leading to stunting of growth (Buller, Carty et al.
2009). Because of this, minocycline is not typically used in neonatal populations. Even though
it is not used clinically in pediatric populations, it is still useful in animal models of neonatal
neuroinflammation. Benefits seen with minocycline administration may help elucidate the
mechanisms behind the disease process and provide more specific targets for future
therapeutic development.
2.12 Summary and Objectives
28
In summary, environmental insults, including infection, during sensitive developmental
periods can alter normal brain development. Human epidemiological studies have linked
childhood infections with increased risk of developmental and psychiatric disorders, but the
exact mechanism behind this is still unknown (Atladottir, Thorsen et al. 2010). Work over the
past three decades has shed light on the cause of these disorders including, autism and
schizophrenia, and have linked neuroimmune activation and dysregulation as risk factors
(Meyer, Feldon et al. 2011). As shown by the background information, activation of microglia
and neuroinflammation have the potential to disrupt normal brain development.
The majority of work looking at the impact of infection and inflammation in early life has
been conducted in rodent models using the acute inflammatory stimuli lipopolysaccharide (LPS)
or Poly I:C (Meyer and Feldon 2010). While many important discoveries about how early life
inflammation affects brain and behavioral have been made in rodent models, additional work in
larger animal species and chronic inflammatory stimuli is needed.
The piglet presents as an ideal animal model for testing hypothesis in the neonatal time
period. The piglet has very similar brain growth and development to humans and has a rich
history of being used as an animal model for neonatal research. There is very little research
being conducted on how chronic immune activation, typical of many infectious agents, can
affect brain development. Here we use a live virus, PRRS, as a model of respiratory syncytium
virus (RSV) in neonates. RSV is a significant clinical burden with over 132,000 children under 5
years of age hospitalized from 1997-2006 (Centers for Disease Control and Prevention 2013).
Neonates are especially prone to severe RSV infection with the median duration of illness of 12
days and 10 percent remaining ill for over four weeks (Steiner 2004). RSV also causes
increases in peripheral pro-inflammatory cytokines (Arnold, Konig et al. 1995). Infection with
PRRS provides us a model to test the short- and long-term effects of peripheral inflammation on
brain development.
29
This dissertation was designed with two objectives. The first was to develop non-
invasive imaging techniques to measure brain growth and development in the piglet. MRI
methods were first developed to measure regional brain volumes and these techniques were
then utilized to characterize the normal brain growth of the domestic pig. The creation of an
averaged brain and MRI-based atlas for the piglet allowed for improvement on MRI analysis
including voxel based morphometry. Methods for additional imaging techniques in the piglet,
including DTI and MR-spec, were also produced.
The second objective was to investigate if peripheral infection with the PRRS virus could
lead to changes in brain development. Using the MRI methods developed, changes to regional
gray and white matter volumes, white matter track structure, and neurochemical concentration
were assessed in piglets inoculated with PRRS. In addition, changes to neurogenesis, cell fate,
and hippocampal neuron morphology were characterized. Finally, the therapeutic potential of
minocycline for reduction of microglia activation and neuroinflammation was conducted in the
PRRS model.
The use of a highly translatable animal model with peripheral viral infection provides
important and clinically relevant information on how peripheral immune activation can change
neurodevelopment. Further research into the mechanisms of these changes may provide novel
therapeutic targets for treatment of neonatal infections. Long-term studies using this model
would also be beneficial to see if changes persist into adulthood.
30
2.13 References
1. Patten, B.M., The embryology of the pig. 2nd ed1931, Philadelphia: P. Blakiston's Son and Co.
2. Book, S.A. and L.K. Bustad, The Fetal and Neonatal Pig in Biomedical Research. J Anim Sci, 1974. 38(5): p. 997-1002.
3. van Straaten, H.W.M., et al., Neurulation in the pig embryo. Anatomy and Embryology,
2000. 202(2): p. 75-84. 4. Nielsen, K.B., et al., Reelin expression during embryonic development of the pig brain.
BMC Neurosci, 2010. 11: p. 75. 5. Dickerson, J.W.T. and J. Dobbing, Prenatal and postnatal growth and development of
the central nervous system of the pig. Proc Biol Sci, 1967. 166(1005): p. 384-395. 6. Pond, W.G., et al., Perinatal ontogeny of brain growth in the domestic pig. Proc Soc Exp
Biol Med, 2000. 223(1): p. 102-8. 7. Dobbing, J. and J. Sands, Comparative aspects of the brain growth spurt. Early Hum
Dev, 1979. 3(1): p. 79-83. 8. Passingham, R.E., Rates of Brain Development in Mammals Including Man. Brain,
Behavior and Evolution, 1985. 26(3-4): p. 167-175. 9. Conrad, M.S., R.N. Dilger, and R.W. Johnson, Brain Growth of the Domestic Pig (Sus
scrofa) from 2 to 24 Weeks of Age: A Longitudinal MRI Study. Dev Neurosci, 2012. 34(4): p. 291-8.
10. Jelsing, J., et al., The postnatal development of neocortical neurons and glial cells in the
Gottingen minipig and the domestic pig brain. J Exp Biol, 2006. 209(Pt 8): p. 1454-62. 11. Lind, N.M., et al., The use of pigs in neuroscience: modeling brain disorders. Neurosci
Biobehav Rev, 2007. 31(5): p. 728-51. 12. Niblock, M.M., et al., Comparative anatomical assessment of the piglet as a model for
the developing human medullary serotonergic system. Brain Res Brain Res Rev, 2005. 50(1): p. 169-83.
13. Rosa-Neto, P., D.J. Doudet, and P. Cumming, Gradients of dopamine D1- and D2/3-
binding sites in the basal ganglia of pig and monkey measured by PET. NeuroImage, 2004. 22(3): p. 1076-83.
14. Larsen, M., et al., The anatomy of the porcine subthalamic nucleus evaluated with
immunohistochemistry and design-based stereology. Anat Embryol (Berl), 2004. 208(3): p. 239-47.
31
15. Agarwal, R.K., et al., Distribution of catecholamines in the central nervous system of the pig. Brain Res Bull, 1993. 32(3): p. 285-91.
16. Anderson, N.J., et al., Characterisation of imidazoline I2 binding sites in pig brain. Eur J
Pharmacol, 2005. 519(1-2): p. 68-74. 17. Marcilloux, J.C., et al., Preliminary results of a magnetic resonance imaging (MRI) study
of the pig brain placed in stereotaxic conditions. Neurosci Lett, 1993. 156(1-2): p. 113-6. 18. Sørensen, J.C., et al., Oriented sectioning of irregular tissue blocks in relation to
computerized scanning modalities:: Results from the domestic pig brain. Journal of Neuroscience Methods, 2000. 104(1): p. 93-98.
19. Felix, B., et al., Stereotaxic atlas of the pig brain. Brain Res Bull, 1999. 49(1-2): p. 1-137. 20. Watanabe, H., et al., MR-based statistical atlas of the Gottingen minipig brain.
Neuroimage, 2001. 14(5): p. 1089-96. 21. Rosa-Neto, P., et al., MDMA-evoked changes in [11C]raclopride and [11C]NMSP
binding in living pig brain. Synapse, 2004. 53(4): p. 222-233. 22. Saikali, S., et al., A three-dimensional digital segmented and deformable brain atlas of
the domestic pig. J Neurosci Meth, 2010. 192(1): p. 102-9. 23. Conrad, M.S., et al., Magnetic resonance imaging of the neonatal piglet brain. Pediatr
Res, 2012. 71(2): p. 179-184. 24. Winter, J.D., et al., Noninvasive MRI measures of microstructural and cerebrovascular
changes during normal swine brain development. Pediatr Res, 2011. 69(5 Pt 1): p. 418-24.
25. Alexander, A.L., et al., Diffusion tensor imaging of the brain. Neurotherapeutics, 2007.
4(3): p. 316-29. 26. Thornton, J.S., et al., Anisotropic water diffusion in white and gray matter of the neonatal
piglet brain before and after transient hypoxia-ischaemia. Magnetic Resonance Imaging, 1997. 15(4): p. 433-440.
27. Blüml, S., Magnetic Resonance Spectroscopy: Basics, in MR Spectroscopy of Pediatric
Brain Disorders, S. Blüml and A. Panigrahy, Editors. 2013, Springer New York. p. 11-23. 28. Smith, D.H., et al., Magnetic resonance spectroscopy of diffuse brain trauma in the pig. J
Neurotrauma, 1998. 15(9): p. 665-74. 29. Munkeby, B.H., et al., A piglet model for detection of hypoxic-ischemic brain injury with
magnetic resonance imaging. Acta Radiol, 2008. 49(9): p. 1049-57. 30. Fang, M., et al., fMRI mapping of cortical centers following visual stimulation in postnatal
pigs of different ages. Life sciences, 2006. 78(11): p. 1197-201.
32
31. Fang, M., et al., Postnatal changes in functional activities of the pig's brain: a combined functional magnetic resonance imaging and immunohistochemical study. Neurosignals, 2005. 14(5): p. 222-33.
32. Workman, A.D., et al., Modeling transformations of neurodevelopmental sequences
across mammalian species. J Neurosci, 2013. 33(17): p. 7368-83. 33. Meurens, F., et al., The pig: a model for human infectious diseases. Trends in
microbiology, 2012. 20(1): p. 50-57. 34. Freeman, T., et al., A gene expression atlas of the domestic pig. BMC Biology, 2012.
10(1): p. 1-22. 35. Dawson, H.D., A Comparative Assessment of the Pig, Mouse and Human Genomes, in
The Minipig in Biomedical Research2011, CRC Press. p. 323-342. 36. Pampusch, M.S., et al., Inducible nitric oxide synthase expression in porcine immune
cells. Vet Immunol Immunopathol, 1998. 61(2-4): p. 279-89. 37. Svedman, P., et al., Staphylococcal wound infection in the pig: Part I. Course. Ann Plast
Surg, 1989. 23(3): p. 212-8. 38. Jensen, H.E., et al., A non-traumatic Staphylococcus aureus osteomyelitis model in pigs.
In Vivo, 2010. 24(3): p. 257-64. 39. Luna, C.M., et al., Animal models of ventilator-associated pneumonia. Eur Respir J,
2009. 33(1): p. 182-8. 40. Elahi, S., et al., Maternal immunity provides protection against pertussis in newborn
piglets. Infect Immun, 2006. 74(5): p. 2619-27. 41. Saif, L.J., et al., The gnotobiotic piglet as a model for studies of disease pathogenesis
and immunity to human rotaviruses. Arch Virol Suppl, 1996. 12: p. 153-61. 42. Roohey, T., T.N. Raju, and A.N. Moustogiannis, Animal models for the study of perinatal
hypoxic-ischemic encephalopathy: a critical analysis. Early Hum Dev, 1997. 47(2): p. 115-46.
43. Raju, T.N., Some animal models for the study of perinatal asphyxia. Biol Neonate, 1992.
62(4): p. 202-14. 44. Haaland, K., et al., Posthypoxic hypothermia in newborn piglets. Pediatr Res, 1997. 41(4
Pt 1): p. 505-12. 45. Bjorkman, S.T., et al., Hypoxic/Ischemic models in newborn piglet: comparison of
constant FiO2 versus variable FiO2 delivery. Brain Res, 2006. 1100(1): p. 110-7. 46. Agnew, D.M., et al., Hypothermia for 24 hours after asphyxic cardiac arrest in piglets
provides striatal neuroprotection that is sustained 10 days after rewarming. Pediatr Res, 2003. 54(2): p. 253-62.
33
47. Alonso-Spilsbury, M., et al., Perinatal asphyxia pathophysiology in pig and human: a review. Anim Reprod Sci, 2005. 90(1-2): p. 1-30.
48. LeBlanc, M.H., et al., MK-801 does not protect against hypoxic-ischemic brain injury in
piglets. Stroke, 1991. 22(10): p. 1270-5. 49. Barkovich, A.J., et al., MR imaging, MR spectroscopy, and diffusion tensor imaging of
sequential studies in neonates with encephalopathy. AJNR Am J Neuroradiol, 2006. 27(3): p. 533-47.
50. Vial, F., et al., A newborn piglet study of moderate hypoxic-ischemic brain injury by 1H-
MRS and MRI. Magn Reson Imaging, 2004. 22(4): p. 457-65. 51. Cheng, Y., et al., Early diffusion weighted imaging and expression of heat shock protein
70 in newborn pigs with hypoxic ischaemic encephalopathy. Postgraduate Medical Journal, 2005. 81(959): p. 589-593.
52. Munkeby, B.H., et al., Morphological and hemodynamic magnetic resonance
assessment of early neonatal brain injury in a piglet model. J Magn Reson Imaging, 2004. 20(1): p. 8-15.
53. Higgins, R.D., et al., Hypothermia and Other Treatment Options for Neonatal
Encephalopathy: An Executive Summary of the Eunice Kennedy Shriver NICHD Workshop. The Journal of Pediatrics, 2011. 159(5): p. 851-858.e1.
54. Thoresen, M., et al., A piglet survival model of posthypoxic encephalopathy. Pediatr Res,
1996. 40(5): p. 738-48. 55. Thoresen, M., et al., Time is brain: starting therapeutic hypothermia within three hours
after birth improves motor outcome in asphyxiated newborns. Neonatology, 2013. 104(3): p. 228-33.
56. Tagin, M.A., et al., Hypothermia for neonatal hypoxic ischemic encephalopathy: an
updated systematic review and meta-analysis. Arch Pediatr Adolesc Med, 2012. 166(6): p. 558-66.
57. Shankaran, S., et al., Childhood outcomes after hypothermia for neonatal
encephalopathy. N Engl J Med, 2012. 366(22): p. 2085-92. 58. Perlman, J.M., et al., Part 11: Neonatal resuscitation: 2010 International Consensus on
Cardiopulmonary Resuscitation and Emergency Cardiovascular Care Science With Treatment Recommendations. Circulation, 2010. 122(16 Suppl 2): p. S516-38.
59. Zanelli, S.A., et al., Implementation of a 'Hypothermia for HIE' program: 2-year
experience in a single NICU. J Perinatol, 2008. 28(3): p. 171-5. 60. Barron, K.D., The microglial cell. A historical review. Journal of the Neurological
Sciences, 1995. 134, Supplement(0): p. 57-68. 61. Lawson, L.J., et al., Heterogeneity in the distribution and morphology of microglia in the
normal adult mouse brain. Neuroscience, 1990. 39(1): p. 151-70.
34
62. Mittelbronn, M., et al., Local distribution of microglia in the normal adult human central nervous system differs by up to one order of magnitude. Acta Neuropathol, 2001. 101(3): p. 249-55.
63. Ginhoux, F., et al., Fate Mapping Analysis Reveals That Adult Microglia Derive from
Primitive Macrophages. Science, 2010. 330(6005): p. 841-845. 64. Saijo, K. and C.K. Glass, Microglial cell origin and phenotypes in health and disease. Nat
Rev Immunol, 2011. 11(11): p. 775-787. 65. Schwarz, J.M., P.W. Sholar, and S.D. Bilbo, Sex differences in microglial colonization of
the developing rat brain. Journal of Neurochemistry, 2012. 120(6): p. 948-963. 66. Hanisch, U.-K. and H. Kettenmann, Microglia: active sensor and versatile effector cells in
the normal and pathologic brain. Nat Neurosci, 2007. 10(11): p. 1387-1394. 67. Nimmerjahn, A., F. Kirchhoff, and F. Helmchen, Resting Microglial Cells Are Highly
Dynamic Surveillants of Brain Parenchyma in Vivo. Science, 2005. 308(5726): p. 1314-1318.
68. Ransohoff, R.M. and V.H. Perry, Microglial physiology: unique stimuli, specialized
responses. Annu Rev Immunol, 2009. 27: p. 119-45. 69. Michelucci, A., et al., Characterization of the microglial phenotype under specific pro-
inflammatory and anti-inflammatory conditions: Effects of oligomeric and fibrillar amyloid-beta. J Neuroimmunol, 2009. 210(1-2): p. 3-12.
70. Perry, V.H. and S. Gordon, Macrophages and microglia in the nervous system. Trends
Neurosci, 1988. 11(6): p. 273-7. 71. Gordon, S., Alternative activation of macrophages. Nat Rev Immunol, 2003. 3(1): p. 23-
35. 72. Song, E., et al., Influence of alternatively and classically activated macrophages on
fibrogenic activities of human fibroblasts. Cell Immunol, 2000. 204(1): p. 19-28. 73. Bilbo, S.D. and J.M. Schwarz, Early-life programming of later-life brain and behavior: a
critical role for the immune system. Front Behav Neurosci, 2009. 3: p. 14. 74. Cuadros, M.A. and J. Navascues, The origin and differentiation of microglial cells during
development. Prog Neurobiol, 1998. 56(2): p. 173-89. 75. Harry, G.J., Microglia during development and aging. Pharmacol Ther, 2013. 139(3): p.
313-26. 76. Kim, S.U. and J. de Vellis, Microglia in health and disease. J Neurosci Res, 2005. 81(3):
p. 302-13. 77. Rochefort, N., et al., Microglia and astrocytes may participate in the shaping of visual
callosal projections during postnatal development. J Physiol Paris, 2002. 96(3-4): p. 183-92.
35
78. Tremblay, M.-È., R.L. Lowery, and A.K. Majewska, Microglial Interactions with Synapses Are Modulated by Visual Experience. PLoS Biol, 2010. 8(11): p. e1000527.
79. Paolicelli, R.C., et al., Synaptic Pruning by Microglia Is Necessary for Normal Brain
Development. Science, 2011. 80. Wakselman, S., et al., Developmental neuronal death in hippocampus requires the
microglial CD11b integrin and DAP12 immunoreceptor. J Neurosci, 2008. 28(32): p. 8138-43.
81. Rigato, C., et al., Pattern of invasion of the embryonic mouse spinal cord by microglial
cells at the time of the onset of functional neuronal networks. Glia, 2011. 59(4): p. 675-95.
82. Aarum, J., et al., Migration and differentiation of neural precursor cells can be directed
by microglia. Proc Natl Acad Sci U S A, 2003. 100(26): p. 15983-8. 83. Jonakait, G.M., et al., Macrophage cell-conditioned medium promotes cholinergic
differentiation of undifferentiated progenitors and synergizes with nerve growth factor action in the developing basal forebrain. Exp Neurol, 2000. 161(1): p. 285-96.
84. Antony, J.M., et al., Endogenous microglia regulate development of embryonic cortical
precursor cells. J Neurosci Res, 2011. 89(3): p. 286-98. 85. Dantzer, R., et al., From inflammation to sickness and depression: when the immune
system subjugates the brain. Nat Rev Neurosci, 2008. 9(1): p. 46-56. 86. Sawada, M., et al., Expression of cytokine receptors in cultured neuronal and glial cells.
Neurosci Lett, 1993. 160(2): p. 131-4. 87. Dantzer, R., Cytokine, sickness behavior, and depression. Immunol Allergy Clin North
Am, 2009. 29(2): p. 247-64. 88. Kumar, H., T. Kawai, and S. Akira, Pathogen recognition by the innate immune system.
Int Rev Immunol, 2011. 30(1): p. 16-34. 89. Pugazhenthi, S., et al., Induction of an Inflammatory Loop by Interleukin-1β and Tumor
Necrosis Factor-α Involves NF-kB and STAT-1 in Differentiated Human Neuroprogenitor Cells. PLoS One, 2013. 8(7): p. e69585.
90. Dantzer, R., et al., Neural and humoral pathways of communication from the immune
system to the brain: parallel or convergent? Auton Neurosci, 2000. 85(1-3): p. 60-5. 91. Dantzer, R. and K.W. Kelley, Twenty years of research on cytokine-induced sickness
behavior. Brain, Behavior, and Immunity, 2007. 21(2): p. 153-160. 92. Kelley, K.W., et al., Cytokine-induced sickness behavior. Brain Behav Immun, 2003. 17
Suppl 1: p. S112-8.
36
93. Johnson, R.W., The concept of sickness behavior: a brief chronological account of four key discoveries. Veterinary Immunology and Immunopathology, 2002. 87(3-4): p. 443-450.
94. Conti, B., et al., Cytokines and fever. Front Biosci, 2004. 9: p. 1433-49. 95. Wan, W., et al., Differential induction of c-Fos immunoreactivity in hypothalamus and
brain stem nuclei following central and peripheral administration of endotoxin. Brain Res Bull, 1993. 32(6): p. 581-7.
96. Goehler, L.E., et al., Interleukin-1 induces c-Fos immunoreactivity in primary afferent
neurons of the vagus nerve. Brain Res, 1998. 804(2): p. 306-10. 97. Kapcala, L.P., et al., Subdiaphragmatic vagotomy inhibits intra-abdominal interleukin-1
beta stimulation of adrenocorticotropin secretion. Brain Res, 1996. 728(2): p. 247-54. 98. Romanovsky, A.A., et al., The vagus nerve in the thermoregulatory response to systemic
inflammation. Am J Physiol, 1997. 273(1 Pt 2): p. R407-13. 99. Banks, W.A., S.A. Farr, and J.E. Morley, Entry of blood-borne cytokines into the central
nervous system: effects on cognitive processes. Neuroimmunomodulation, 2002. 10(6): p. 319-27.
100. Van Dam, A.M., et al., Interleukin-1 receptors on rat brain endothelial cells: a role in
neuroimmune interaction? FASEB J, 1996. 10(2): p. 351-6. 101. Done, S.H., D.J. Paton, and M.E. White, Porcine reproductive and respiratory syndrome
(PRRS): a review, with emphasis on pathological, virological and diagnostic aspects. Br Vet J, 1996. 152(2): p. 153-74.
102. Dokland, T., The structural biology of PRRSV. Virus Research, 2010. 154(1-2): p. 86-97. 103. Rossow, K.D., et al., Porcine reproductive and respiratory syndrome virus infection in
neonatal pigs characterised by marked neurovirulence. Vet Rec., 1999. 144(16): p. 444-448.
104. Holtkamp, D.J., J.B. Kliebenstein, and E.J. Neumann, Assessment of the economic
impact of porcine reproductive and respiratory syndrome virus on United States pork producers. Journal of Swine Health and Production, 2013. 21(2): p. 72-84.
105. Amonsin, A., et al., Comparative analysis of complete nucleotide sequence of porcine
reproductive and respiratory syndrome virus (PRRSV) isolates in Thailand (US and EU genotypes). Virol J, 2009. 6: p. 143.
106. Van Breedam, W., et al., Porcine reproductive and respiratory syndrome virus entry into
the porcine macrophage. Journal of General Virology, 2010. 91(7): p. 1659-1667. 107. Van Breedam, W., et al., The M/GP(5) glycoprotein complex of porcine reproductive and
respiratory syndrome virus binds the sialoadhesin receptor in a sialic acid-dependent manner. PLoS Pathog, 2010. 6(1): p. e1000730.
37
108. Calvert, J.G., et al., CD163 expression confers susceptibility to porcine reproductive and respiratory syndrome viruses. Journal of virology, 2007. 81(14): p. 7371-9.
109. Zhang, C., et al., Profiling of cellular proteins in porcine reproductive and respiratory
syndrome virus virions by proteomics analysis. Virol J, 2010. 7: p. 242. 110. Zhang, L., et al., Comparative expression of Toll-like receptors and inflammatory
cytokines in pigs infected with different virulent porcine reproductive and respiratory syndrome virus isolates. Virol J, 2013. 10(1): p. 135.
111. Miguel, J.C., et al., Expression of inflammatory cytokines and Toll-like receptors in the
brain and respiratory tract of pigs infected with porcine reproductive and respiratory syndrome virus. Veterinary Immunology and Immunopathology, 2010. 135(3-4): p. 314-319.
112. Elmore, M.R., et al., Respiratory viral infection in neonatal piglets causes marked
microglia activation in the hippocampus and deficits in spatial learning. J Neurosci, 2014. 34(6): p. 2120-9.
113. Lieberman, A.P., et al., Production of tumor necrosis factor and other cytokines by
astrocytes stimulated with lipopolysaccharide or a neurotropic virus. Proc Natl Acad Sci U S A, 1989. 86(16): p. 6348-52.
114. Kigerl, K.A., et al., Identification of two distinct macrophage subsets with divergent
effects causing either neurotoxicity or regeneration in the injured mouse spinal cord. J Neurosci, 2009. 29(43): p. 13435-44.
115. Olah, M., et al., Microglia phenotype diversity. CNS Neurol Disord Drug Targets, 2011.
10(1): p. 108-18. 116. Yirmiya, R. and I. Goshen, Immune modulation of learning, memory, neural plasticity
and neurogenesis. Brain, Behavior, and Immunity, 2011. 25(2): p. 181-213. 117. Rice, D. and S. Barone, Jr., Critical periods of vulnerability for the developing nervous
system: evidence from humans and animal models. Environ Health Perspect, 2000. 108 Suppl 3: p. 511-33.
118. Ekdahl, C.T., Microglial activation - tuning and pruning adult neurogenesis. Frontiers in
Pharmacology, 2012. 3. 119. Kempermann, G., et al., Early determination and long-term persistence of adult-
generated new neurons in the hippocampus of mice. Development, 2003. 130(2): p. 391-9.
120. Biebl, M., et al., Analysis of neurogenesis and programmed cell death reveals a self-
renewing capacity in the adult rat brain. Neurosci Lett, 2000. 291(1): p. 17-20. 121. Nakashiba, T., et al., Young dentate granule cells mediate pattern separation, whereas
old granule cells facilitate pattern completion. Cell, 2012. 149(1): p. 188-201.
38
122. Belarbi, K. and S. Rosi, Modulation of adult-born neurons in the inflamed hippocampus. Front Cell Neurosci, 2013. 7: p. 145.
123. Kohman, R.A. and J.S. Rhodes, Neurogenesis, inflammation and behavior. Brain Behav
Immun, 2013. 27C: p. 22-32. 124. Monje, M.L., H. Toda, and T.D. Palmer, Inflammatory Blockade Restores Adult
Hippocampal Neurogenesis. Science, 2003. 302(5651): p. 1760-1765. 125. Ekdahl, C.T., et al., Inflammation is detrimental for neurogenesis in adult brain.
Proceedings of the National Academy of Sciences, 2003. 100(23): p. 13632-13637. 126. Cacci, E., et al., In vitro neuronal and glial differentiation from embryonic or adult neural
precursor cells are differently affected by chronic or acute activation of microglia. Glia, 2008. 56(4): p. 412-425.
127. Wu, M.D., et al., Adult murine hippocampal neurogenesis is inhibited by sustained IL-
1beta and not rescued by voluntary running. Brain Behav Immun, 2012. 26(2): p. 292-300.
128. Wu, M.D., et al., Sustained IL-1β expression impairs adult hippocampal neurogenesis
independent of IL-1 signaling in nestin+ neural precursor cells. Brain, Behavior, and Immunity, 2013. 32(0): p. 9-18.
129. Vallieres, L., et al., Reduced hippocampal neurogenesis in adult transgenic mice with
chronic astrocytic production of interleukin-6. J Neurosci, 2002. 22(2): p. 486-92. 130. Fujioka, H. and T. Akema, Lipopolysaccharide acutely inhibits proliferation of neural
precursor cells in the dentate gyrus in adult rats. Brain Research, 2010. 1352(0): p. 35-42.
131. Kiyota, T., et al., AAV serotype 2/1-mediated gene delivery of anti-inflammatory
interleukin-10 enhances neurogenesis and cognitive function in APP+PS1 mice. Gene Ther, 2012. 19(7): p. 724-33.
132. Kiyota, T., et al., CNS expression of anti-inflammatory cytokine interleukin-4 attenuates
Alzheimer's disease-like pathogenesis in APP+PS1 bigenic mice. FASEB J, 2010. 24(8): p. 3093-102.
133. Battista, D., et al., Neurogenic niche modulation by activated microglia: transforming
growth factor β increases neurogenesis in the adult dentate gyrus. European Journal of Neuroscience, 2006. 23(1): p. 83-93.
134. Aberg, M.A., et al., Peripheral infusion of IGF-I selectively induces neurogenesis in the
adult rat hippocampus. J Neurosci, 2000. 20(8): p. 2896-903. 135. Stolp, H.B., Neuropoietic cytokines in normal brain development and
neurodevelopmental disorders. Molecular and Cellular Neuroscience, 2013. 53(0): p. 63-68.
39
136. Green, H.F., et al., A role for interleukin-1beta in determining the lineage fate of embryonic rat hippocampal neural precursor cells. Mol Cell Neurosci, 2012. 49(3): p. 311-21.
137. Jakubs, K., et al., Inflammation regulates functional integration of neurons born in adult
brain. J Neurosci, 2008. 28(47): p. 12477-88. 138. Jarlestedt, K., et al., Decreased survival of newborn neurons in the dorsal hippocampus
after neonatal LPS exposure in mice. Neuroscience, 2013. 253(0): p. 21-8. 139. Cui, K., et al., Effects of prenatal immune activation on hippocampal neurogenesis in the
rat. Schizophrenia Research, 2009. 113(2-3): p. 288-297. 140. Jiang, P., et al., The persistent effects of maternal infection on the offspring's cognitive
performance and rates of hippocampal neurogenesis. Prog Neuropsychopharmacol Biol Psychiatry, 2013. 44(0): p. 279-89.
141. von Bohlen Und Halbach, O., Structure and function of dendritic spines within the
hippocampus. Ann Anat, 2009. 191(6): p. 518-31. 142. Olton, D.S., J.T. Becker, and G.E. Handelmann, Hippocampus, space, and memory.
Behavioral and Brain Sciences, 1979. 2(03): p. 313-322. 143. Green, E.J. and J.M. Juraska, The dendritic morphology of hippocampal dentate granule
cells varies with their position in the granule cell layer: a quantitative Golgi study. Experimental Brain Research, 1985. 59(3): p. 582-586.
144. Juraska, J.M., et al., Sex differences in the dendritic branching of dentate granule cells
following differential experience. Brain Research, 1985. 333(1): p. 73-80. 145. McEwen, B.S., Stress and hippocampal plasticity. Annu Rev Neurosci, 1999. 22: p. 105-
22. 146. Woolley, C.S., E. Gould, and B.S. McEwen, Exposure to excess glucocorticoids alters
dendritic morphology of adult hippocampal pyramidal neurons. Brain Res, 1990. 531(1-2): p. 225-31.
147. Richwine, A.F., et al., Architectural changes to CA1 pyramidal neurons in adult and aged
mice after peripheral immune stimulation. Psychoneuroendocrinology, 2008. 33(10): p. 1369-77.
148. Milatovic, D., et al., Pharmacologic suppression of neuronal oxidative damage and
dendritic degeneration following direct activation of glial innate immunity in mouse cerebrum. J Neurochem, 2003. 87(6): p. 1518-26.
149. Jurgens, H.A., K. Amancherla, and R.W. Johnson, Influenza Infection Induces
Neuroinflammation, Alters Hippocampal Neuron Morphology, and Impairs Cognition in Adult Mice. The Journal of Neuroscience, 2012. 32(12): p. 3958-3968.
40
150. Baharnoori, M., W.G. Brake, and L.K. Srivastava, Prenatal immune challenge induces developmental changes in the morphology of pyramidal neurons of the prefrontal cortex and hippocampus in rats. Schizophr Res, 2009. 107(1): p. 99-109.
151. Pickett, J. and E. London, The neuropathology of autism: a review. J Neuropathol Exp
Neurol, 2005. 64(11): p. 925-35. 152. Kolluri, N., et al., Lamina-specific reductions in dendritic spine density in the prefrontal
cortex of subjects with schizophrenia. Am J Psychiatry, 2005. 162(6): p. 1200-2. 153. Glantz, L.A. and D.A. Lewis, Decreased dendritic spine density on prefrontal cortical
pyramidal neurons in schizophrenia. Arch Gen Psychiatry, 2000. 57(1): p. 65-73. 154. Bitzer-Quintero, O.K. and I. Gonzalez-Burgos, Immune system in the brain: a modulatory
role on dendritic spine morphophysiology? Neural Plast, 2012. 2012: p. 348642. 155. Kondo, S., S. Kohsaka, and S. Okabe, Long-term changes of spine dynamics and
microglia after transient peripheral immune response triggered by LPS in vivo. Mol Brain, 2011. 4: p. 27.
156. Neves, G., S.F. Cooke, and T.V. Bliss, Synaptic plasticity, memory and the
hippocampus: a neural network approach to causality. Nat Rev Neurosci, 2008. 9(1): p. 65-75.
157. Kasai, H., et al., Structural dynamics of dendritic spines in memory and cognition.
Trends Neurosci, 2010. 33(3): p. 121-9. 158. Khandaker, G.M., et al., Childhood infection and adult schizophrenia: a meta-analysis of
population-based studies. Schizophr Res, 2012. 139(1-3): p. 161-8. 159. Liu, F. and L.D. McCullough, Inflammatory responses in hypoxic ischemic
encephalopathy. Acta Pharmacol Sin, 2013. 34(9): p. 1121-30. 160. Stirling, D.P., et al., Minocycline as a neuroprotective agent. Neuroscientist, 2005. 11(4):
p. 308-22. 161. He, Y., S. Appel, and W. Le, Minocycline inhibits microglial activation and protects nigral
cells after 6-hydroxydopamine injection into mouse striatum. Brain Res, 2001. 909(1-2): p. 187-93.
162. Guo, G. and N.R. Bhat, p38alpha MAP kinase mediates hypoxia-induced motor neuron
cell death: a potential target of minocycline's neuroprotective action. Neurochem Res, 2007. 32(12): p. 2160-6.
163. Walton, K.M., et al., Activation of p38MAPK in Microglia After Ischemia. Journal of
Neurochemistry, 1998. 70(4): p. 1764-1767. 164. Kim, H.S. and Y.H. Suh, Minocycline and neurodegenerative diseases. Behav Brain
Res, 2009. 196(2): p. 168-79.
41
165. Chen, M., et al., Minocycline inhibits caspase-1 and caspase-3 expression and delays mortality in a transgenic mouse model of Huntington disease. Nature medicine, 2000. 6(7): p. 797-801.
166. Wang, J., et al., Minocycline up-regulates Bcl-2 and protects against cell death in
mitochondria. J Biol Chem, 2004. 279(19): p. 19948-54. 167. Fan, L.W., et al., Minocycline attenuates lipopolysaccharide-induced white matter injury
in the neonatal rat brain. Neuroscience, 2005. 133(1): p. 159-68. 168. Tsuji, M., et al., Minocycline worsens hypoxic-ischemic brain injury in a neonatal mouse
model. Exp Neurol, 2004. 189(1): p. 58-65. 169. Buller, K.M., et al., Minocycline: A neuroprotective agent for hypoxic-ischemic brain
injury in the neonate? Journal of Neuroscience Research, 2009. 87(3): p. 599-608. 170. Gordon, P.H., et al., Efficacy of minocycline in patients with amyotrophic lateral
sclerosis: a phase III randomised trial. The Lancet Neurology, 2007. 6(12): p. 1045-1053.
171. Chaudhry, I.B., et al., Minocycline benefits negative symptoms in early schizophrenia: a
randomised double-blind placebo-controlled clinical trial in patients on standard treatment. Journal of Psychopharmacology, 2012. 26(9): p. 1185-1193.
172. Atladottir, H.O., et al., Association of Hospitalization for Infection in Childhood With
Diagnosis of Autism Spectrum Disorders: A Danish Cohort Study. Arch Pediatr Adolesc Med, 2010. 164(5): p. 470-477.
173. Meyer, U., J. Feldon, and O. Dammann, Schizophrenia and autism: both shared and
disorder-specific pathogenesis via perinatal inflammation? Pediatr Res, 2011. 69(5 Pt 2): p. 26R-33R.
174. Meyer, U. and J. Feldon, Epidemiology-driven neurodevelopmental animal models of
schizophrenia. Prog Neurobiol, 2010. 90(3): p. 285-326. 175. Centers for Disease Control and Prevention, Respiratory Syncytial Virus Activity - United
States, July 2011- January 2013. MMWR, 2013. 62(08): p. 141-144. 176. Steiner, R.W., Treating acute bronchiolitis associated with RSV. Am Fam Physician,
2004. 69(2): p. 325-30. 177. Arnold, R., et al., Cytokine (IL-8, IL-6, TNF-alpha) and soluble TNF receptor-I release
from human peripheral blood mononuclear cells after respiratory syncytial virus infection. Immunology, 1995. 85(3): p. 364-72.
42
CHAPTER 3
EARLY POSTNATAL RESPIRATORY VIRAL INFECTION INDUCES STRUCTURAL AND NEUROCHEMICAL CHANGES IN THE NEONATAL PIGLET BRAIN
3.1 Abstract
Respiratory infections during the neonatal period are common, yet little is known about their
impact on human brain development because studies in infants are either unethical or extremely
difficult. Furthermore, results from rodent models are difficult to translate to human infants due
to the differences in brain development and morphology. To overcome some of these barriers,
in the present study domestic piglets that have brain growth and morphology similar to human
infants were inoculated with porcine reproductive and respiratory syndrome virus (PRRSV) on
postnatal day (PD) 7 and magnetic resonance imaging (MRI) was used to assess brain
macrostructure (voxel-based morphometry), microstructure (diffusion tensor imaging) and
biochemistry (MR spectroscopy) at PD 29 or 30. PRRSV piglets exhibited signs of infection
throughout the post inoculation period and had elevated plasma levels of TNFα at the end of the
study, indicating PRRSV caused a sustained inflammatory response. PRRSV infection
increased the volume of several components of the ventricular system including the cerebral
aqueduct, fourth ventricle, and the lateral ventricles. Group comparisons between control and
PRRSV piglets defined 8 areas (≥20 clusters) where PRRSV piglets had less gray matter
volume; 5 areas where PRRSV piglets had less white matter volume; and 4 relatively small
areas where PRRSV piglets had more white matter. Of particular interest was a bilateral
reduction in gray and white matter in the primary visual cortex. PRRSV piglets tended to have
reduced fractional anisotropy (an indicator of white matter development) in the corpus callosum.
Additionally, N-acetylaspartate, creatine, and myo-inositol were decreased in the hippocampus
of PRRSV piglets suggesting disrupted neuronal and glial health and energy imbalances.
These finding show that early-life infection can affect brain growth and development.
43
3.2 Introduction
Infectious disease is the most common cause of illness in children, with acute respiratory
infection constituting the most prevalent reason for hospitalization in children under one year of
age (Hall, Weinberg et al. 2009). This is a concern for neurodevelopment because the phase of
rapid brain growth that begins in the late prenatal period continues into the early postnatal
period due to dendritic growth, synaptogenesis, and intense glial cell proliferation (Huttenlocher
1979; Dietrich, Bradley et al. 1988; Rice and Barone 2000). During infection, immune-to-brain
signaling pathways activate microglia, causing neuroinflammation (Dantzer and Kelley 2007).
Developing and mature neurons, as well as glia, have numerous pro-inflammatory cytokine
receptors(Sawada, Itoh et al. 1993), and untimely inflammation in rodent models, especially in
the prenatal period, is known to affect neural development and increase risk for behavioral
disorders that manifest later (Meyer, Feldon et al. 2011). Little is known, however, about
peripheral infection in the neonatal period, especially in humans and other animals whose brain
is gyrencephalic and experience major perinatal growth.
Quantitative magnetic resonance imaging (MRI) can provide important information on
brain development in early childhood and adolescence but because of potential health concerns
for infants, few studies have focused on the period from birth to 2 years of age when dramatic
brain development occurs and progress on understanding the influence of infection on brain
development in infants has been slow. Furthermore, results from rodent models commonly used
to investigate neurodevelopment are difficult to translate to human infants due to the substantial
differences in brain development and morphology. In this regard, the domestic piglet (Sus
scrofa domestica) may be an excellent translational model. Similar to humans, the major brain
growth spurt in pigs extends from the late prenatal to the postnatal period (Dobbing and Sands
1979). Gross anatomical features, including gyral pattern and distribution of gray and white
matter of the neonatal piglet brain are similar to that of human infants (Dickerson and Dobbing
44
1967; Thibault and Margulies 1998). Moreover, their physical size allows neuroimaging
instruments designed for humans to be used with piglets. Indeed, structural MRI, functional
MRI, and positron emission tomography have all been conducted in pigs (Fang, Lorke et al.
2005; Jakobsen, Pedersen et al. 2006; Conrad, Dilger et al. 2012). Additionally, piglets can
undergo cognitive testing at a young age and show far more overlap with humans in genes
involved in immunity compared to rodents (Dilger and Johnson 2010; Dawson 2011; Meurens,
Summerfield et al. 2012). Thus, piglets represent a gyrencephalic species with brain growth
similar to humans that can be used in highly controlled experiments to explore how infection
affects brain structure and function.
In the present study, we used MRI to assess brain structure (voxel-based morphometry),
connectivity (diffusion-tensor imaging) and metabolites (spectroscopy) to test the hypothesis
that infection with porcine reproductive and respiratory syndrome virus (PRRSV) in the early
postnatal period affects brain development in piglets. PRRSV infects mononuclear myeloid
cells in lungs inducing interstitial pneumonia in piglets (Done, Paton et al. 1996). Furthermore,
postnatal PRRSV infection in piglets was recently reported to activate brain microglial cells,
cause neuroinflammation, and reduce performance in a spatial learning and memory task
(Elmore, Burton et al. 2014).
3.3 Materials and Methods
Animals, Housing, and Feeding
Naturally farrowed crossbred piglets from three separate litters (7 males and 7 females)
were obtained from the University of Illinois swine herd. Piglets were brought to the biomedical
animal facility on PD 2 (to allow for colostrum consumption from the sow), where upon arrival,
they were assigned to either the control group (3 males and 3 females) or the PRRSV infection
group (4 males and 4 females) based on sex, litter of origin, and body weight and housed
45
individually in cages (0.76 m L x 0.58 m W x 0.47 m H) designed for neonatal piglets (Elmore,
Burton et al. 2014). Two of the PRRSV females developed severe diarrhea and were removed
from the study. Each cage was positioned in a rack, with stainless steel perforated side walls
and clear acrylic front and rear doors within one of two separate but identical disease
containment chambers that have been described (Elmore, Burton et al. 2014). Each cage was
fitted with flooring designed for neonatal animals (Tenderfoot/NSR, Tandem Products, Inc.,
Minneapolis, MN, USA). A toy (plastic Jingle BallTM, Bio-Serv, Frenchtown, NJ, USA) was
provided to each piglet. Room temperature was maintained at 27ºC and each cage was
equipped with an electric heat pad (K&H Lectro-KennelTM Heat Pad, K&H Manufacturing, LLC,
Colorado Springs, CO, USA). Piglets were maintained on a 12-h light/dark cycle; however,
during the dark cycle minimal lighting was provided.
Piglets were fed a commercial sow milk replacer (Advance Liqui-Wean, Milk Specialties
Co., Dundee, IL, USA). Milk was reconstituted daily to a final concentration of 206 g/L using tap
water and supplied at a rate of 285 mL/kg BW (based on daily recorded weights) to a stainless
steel bowl via a peristaltic pump (Control Company, Friendswood, TX). Using this automated
feeding system (similar to that described previously (Dilger and Johnson 2010)), piglets
received their daily allotted milk over 18 meals (once per hour from 0600 h to 2400 h). All
animal experiments were in accordance with the National Institute of Health Guidelines for the
Care and Use of Laboratory Animals and approved by the University of Illinois at Urbana-
Champaign Institutional Animal Care and Use Committee.
Experimental Design and Assessment of Sickness
At PD 7, piglets were inoculated intranasal with either 1mL of 1x10^5 50% tissue culture
infected dose (TCID 50) of live PRRSV (strain P129-BV, obtained from the School of Veterinary
Medicine at Purdue University, West Lafayette, IN, USA) or sterile phosphate buffered saline
(PBS). Daily body weights (kg) were obtained to monitor piglets’ growth. In addition, daily
46
rectal temperatures were obtained PD 7 through PD 28. The willingness of the piglets to
consume their first daily meal was determined from PD 7 to PD 28 using a feeding score (1 = no
attempt to consume the milk; 2 = attempted to consume the milk, but did not finish within 1 min;
3 = consumed all of the milk within 1 min). Body weight, rectal temperature, and feeding score
data were analyzed as a two-way (treatment x day) repeated measures ANOVA using the
MIXED procedure in SAS (SAS Institute Inc, Cary, NC, USA). There were no significant
differences due to sex, therefore it was not included in final analysis of body weight, rectal
temperature, and feeding score data. Significance was accepted at p<0.05.
The presence of PRRSV antibodies in the serum of all piglets at the end of the study
was analyzed by the Veterinary Diagnostic Laboratory (University of Illinois, Urbana, Illinois)
using a PRRSV-specific ELISA kit (IDEXX Laboratories). This assay has 98.8% sensitivity and
99.9% specificity, with an S/P ratio of >0.4 indicating a positive sample. Serum tumor necrosis
factor alpha (TNF-α) levels at the end of the study were determined using porcine-specific
sandwich enzyme immunoassays (R&D Systems). Data analysis was conducted using the
MIXED procedure in SAS. Serum TNF-α concentrations were analyzed as a two-way
(treatment x sex) ANOVA. Sex was not significant so it was removed from the model and data
were reanalyzed as a one-way ANOVA to determine the effects of treatment. Significance was
accepted at p<0.05.
At PD 29 and PD30 control and PRRSV piglets were transported to the Biomedical
Imaging Center at the Beckman Institute and anesthetized using a telazol:ketamine:xylazine
(100/50/50 mg/kg/BW; Fort Dodge Animal Health). Anesthesia was maintained by inhalation of
isoflurane (98% oxygen/2% isoflurane). A MRI compatible pulse oximeter was used to monitor
heart rate and oxygen saturation throughout the scanning procedure. After piglets entered a
sustainable plane of deep anesthesia, they were placed in the MRI machine. Piglets were
restrained and remained immobilized throughout the MRI scans to prevent motion artifact. After
47
conclusion of the scans, the piglets were euthanized via intracardiac injection of sodium
pentobarbital (Fatal Plus, 72 mg/kg BW).
All MRI was conducted using a Siemens MAGNETOM Trio 3T imager and a 32-channel
head coil (Siemens, Erlangen, Germany). Descriptions of the MRI sequences have been
previously described, and are briefly explained below (Radlowski et al, 2013). The total scan
time per pig was roughly 50 minutes.
Structural MRI Acquisition and Analysis
For brain structure analysis, anatomic images were acquired using a 3D T1-weighted
magnetization-prepared rapid gradient-echo (MPRAGE) sequence. Three repetitions were
acquired using the following parameters: repetition time = 1,900 ms; echo time = 2.49 ms;
inversion time = 900 ms; flip angle = 9°; matrix = 256 x 256, 224 slices with thickness = 0.7 mm.
The final voxel size was 0.7 mm isotropic across the entire head from the tip of the snout to the
cervical/thoracic spinal cord junction as described previously (Conrad, Dilger et al. 2012).
The three MPRAGE image sets were averaged and the brains were manually extracted
using FSL (Analysis Group, FMRIB, Oxford, UK) (Jenkinson, Beckmann et al. 2012). The
extracted brains were brought in to a common stereotactic space by 12 parameter affine
registration to a digital piglet brain atlas (http://pigmri.illinois.edu). The remainder of the voxel-
based morphometry technique was conducted using SPM8 (Wellcome Department of Clinical
Neurology, London; http://www.fil.ion.cul.ac.uk) in MATLAB version 8.0.0.783 (Mathworks). The
brains were segmented into gray matter, white matter, and CSF classifications using the prior
probability maps available in the piglet brain atlas dataset (http://pigmri.illinois.edu). Next, the
Diffeomorphic Anatomical Registration using Exponentiated Lie Algebra (DARTEL) package of
SPM8 was used to create study specific templates (Ashburner 2007). Changes from default
include a bounding box of -30.1 to 30.1, -35 to 44.8, -28 to 31.5; and a voxel size of 0.7 mm3. A
linear bend energy was used as the cost function. Flow fields generated from the DARTEL
48
procedure were converted to warp fields which represent the expansion/shrinkage from the
subject to atlas space. The modulated data were smoothed with a 4 mm full-width half
maximum (FWHM).
Due to the small sample size, the statistical non-parametric methods (SnPM) toolbox
was used to compare the modulated gray and white matter data between the control and
PRRSV groups (http://go.warwick.ac.ui/tenichols/snpm). SnPM uses non-parametric
permutation tests which is advantageous for small numbers per treatment (Nichols and Holmes
2002). A maximum number of permutations were used with a FWHW of 7 mm for variance
smoothing. ANCOVA was used for global normalization and a mean voxel value was used for
Global Calculation. All other options were default. No covariates were used.
Two-sample permutation t tests were performed voxel-wise for both gray and white
matter to detect differences between the control and PRRSV treatments. Due to the small
numbers per treatment, an uncorrected p < 0.01 was used. A threshold of at least 20 edge-
connected voxels (clusters) was set and clusters that appeared on the edges of the brain were
excluded. The differences between control and PRRSV piglets are displayed in the pseudo-t
statistic images (Figures 2 and 3). The location of the significant voxel clusters were estimated
using an MRI-based atlas for adult domestic pigs (Saikali, Meurice et al. 2010).
In addition to the VBM, brain region volume estimations were calculated. Using the warp
fields from the DARTEL procedure, inverse warps were created for each subject. The inverse
warp represents the expansion/shrinkage from the atlas to the subject space. The inverse
warps were then applied to the regions of interest (ROI) maps found in the piglet brain atlas
(http://pigmri.illinois.edu). The volumes were then estimated for each ROI for each subject
using FSLstats. Data analysis was conducted using the MIXED procedure in SAS. ROI
volumes were analyzed as a two-way (treatment x sex) ANOVA. Significance was accepted at
p<0.05. . Unless otherwise stated, data are presented as Least Square Means (LSM) ±
Standard Error of the Means (SEM).
49
Diffusion Tensor Imaging and Analysis
Analysis of neuronal connectivity was conducted using a diffusion tensor imaging (DTI)
sequence to visualize fiber tracts non-invasively. A diffusion-weighted echo-planar-imaging
(DW-EPI) sequence was used with the following parameters: repetition time = 5000 ms; echo
time = 91 ms; averages = 3; diffusion weightings = 2; b-value of 1000 s/mm2 across 30
directions; two images with a b-value of 0 s/mm2. Forty slices with a 2.0 mm thickness were
collected with a matrix size of 100 x 100 for a final voxel size of 2.0 mm isotropic.
The DW-EPI images were then analyzed using the diffusion toolbox in the FSL software
package to create fractional anisotropy (FA) and mean diffusivity (MD) images. The brain was
manually extracted. Masks of the corpus callosum and white matter from the piglet brain atlas
were nonlinearly transformed to the subject’s MPRAGE space and then linearly transformed into
the DTI space. The corpus callosum had a threshold of 0.15 applied to compensate for
expansion caused by interpolation. The white matter ROI had a threshold of 0.5 applied and
was dilated twice. Averages were taken in the regions of interest with a FA>0.2 mask to select
only white matter tracts. Averages were taken over the whole brain with a FA>0.2 mask to
assess global FA values. MD values were averaged over the same region of interest. Data
analysis was conducted using the T-TEST procedure of SAS software. A two sample two-sided
t-test design was used to compare control versus PRRSV animals. Significance was accepted
at p<0.05, statistical trends at p<0.10.
MR Spectroscopy Parameters and Analysis
Magnetic resonance spectroscopy (MRS) was performed using a spin echo chemical
shift sequence with the following parameters: repetition time = 3000 ms; echo time = 30 ms; 128
averages. The voxel size was 12 x 25 x 12 mm and was placed over the left and right dorsal
hippocampus. Six presaturation bands were placed around the volume of interest. Automatic
50
and/or higher-order shims were conducted before each spectroscopic scan to ensure a FWHM
of less than 18 Hz. Both water-suppressed and non-water suppressed data were collected.
All MRS data were subsequently analyzed by LC Model 6.3 fitting program
(Provencher). Absolute levels, in institutional units, were obtained by using eddy current
correction procedure and water scaling. Water-suppressed time domain data were analyzed
between 0.2 and 4.0 ppm without further T1 or T2 correction. There was no post-suppression of
water while using the LC Model. The basis-set of metabolite phantom spectra was used as
prior knowledge; the impact of lack of post-suppression was minimal. Absolute concentrations
were calculated for each metabolite. Cramer-Rao lower bounds (S.D.%) were calculated by the
LC Model to evaluate the quantitative result of absolute levels. Only those levels that were less
than 20% were considered reliable and were included for further statistical analysis. Data
analysis was conducted using the MIXED procedure in SAS. Metabolite concentrations were
analyzed as a two-way (treatment x sex) ANOVA. Sex was not a significant main effect for any
metabolite so it was removed from the model and data were reanalyzed as a one-way ANOVA
to determine the effects of treatment. Significance was accepted at p<0.05.
3.4 Results
PRRSV Infection and Measures of Sickness
Serum ELISA results indicated all control piglets were negative for PRRSV at the end of
the study, whereas all piglets inoculated were positive for PRRSV. Body weight, rectal
temperature, and the willingness to consume the first daily meal (feeding score) were
determined to provide an indication of the sickness response of piglets infected with PRRSV
(Figure 3.1). Analysis of body weight data showed a significant effect of day (F (28, 275) =
31.51, p < 0.001), but neither treatment (F (1, 10) = 0.53) nor the day × treatment interaction (F
51
(27, 275) = 1.25) were significant indicating control and PRRSV piglets gained similar weight
over the course of the study. Nonetheless, analysis of feeding score data revealed a significant
effect of treatment (F (1, 10) = 10.21, p < 0.01), day (F (22, 214) = 2804.45, p < 0.001), and a
trend toward a treatment × day interaction (F (21, 214) = 1.58, p = 0.056), indicating PRRSV
piglets’ motivation to consume the first meal of the day was reduced towards the last half of the
study. It is noteworthy that this is the timeframe when body weight of control and PRRSV
piglets began to separate. Analysis of rectal temperature data showed a significant effect of
treatment (F (1, 10) = 43.19), day (F (22, 215) = 3.56), and a treatment × day interaction (F (21,
215) = 2.88) (all, p < 0.0001). PRRSV piglets became febrile 3 d after inoculation and remained
so during most of the experimental period. The proinflammatory cytokine TNF-α was
significantly increased in PRRSV piglets (F (1, 20) = 98.87 p < 0.001) with an average serum
concentration of 135.6 ± 13.0 pg/mL compared to 17.1 ± 1.7 pg/mL in controls. Collectively,
these data indicate infection with PRRSV in the neonatal period induced clinical signs of illness.
Brain Region Volume Estimation
Using the deformation maps created during the DARTEL procedure, the regions of
interest (ROI) from the piglet brain atlas (http://pigmri.illinois.edu) were warped to the individual
subject’s brain. Volume estimations were made for 19 regions of interest (Table 3.1). In
addition, gray and white matter volumes were estimated by the segmented data from the
DARTEL procedure and whole brain volume was estimated using the extracted brain. Overall,
there were very few differences in brain region volumes between control and PRRSV piglets or
between male and female piglets (Table 3.1). An exception was that PRRSV infection
increased the volume of several components of the ventricular system including the cerebral
52
aqueduct, fourth ventricle, and the lateral ventricles. In the case of the cerebral aqueduct and
fourth ventricle, the effects of PRRSV were greater in females compared to males.
Voxel-based Morphometry
Although there were few effects of PRRSV on brain region volumes, the VBM analysis
(Table 3.2) revealed eight clusters where PRRSV piglets had significantly less gray matter
compared to control piglets. The local maxima coordinates were mapped to the piglet brain
atlas and histological atlas to estimate the anatomic region (http://pigmri.illinois.edu) (Felix,
Leger et al. 1999). Of particular interest are two clusters located in the left and right primary
visual cortex which had the highest pseudo-t value. Pseudo-t maps for areas that have reduced
gray matter volume in PRRSV piglets are shown in Figure 3.2. There were no areas where the
PRRSV piglets had increased gray matter volumes compared to controls.
Analysis of white matter volumes revealed four regions where PRRSV piglets had
increased white matter compared to controls (Table 3.2). These regions had smaller clusters
and lower pseudo-t values compared to clusters where controls had more white matter volume.
There were five clusters where PRRSV piglets had less white matter volume than controls
(Table 3.2). Pseudo-t maps for areas that have reduced white matter volume in PRRSV piglets
are shown in Figure 3.3.
Diffusion Tensor Imaging
Fractional anisotropy maps were compared between control and PRRSV piglets to
evaluate white matter tract differences. An example FA map is shown in Figure 3.4. Three
areas of interest were evaluated for FA and MD differences including whole brain, corpus
53
callosum, and white matter (Figure 3.4). There were no differences in whole brain and white
matter FA (p > 0.10 in both cases). However, when the region of interest was restricted to the
corpus callosum, control piglets tended to have a larger FA than PRRSV piglets (p = 0.06).
There were no significant differences in whole brain, corpus callosum, or white matter for MD (p
> 0.10 in each case).
Magnetic Resonance Spectroscopy
Differences in absolute metabolite concentrations between control and PRRSV piglets
were evaluated in the hippocampus using MRS. The hippocampus was chosen because
PRRSV infection (a) activates microglia in this brain region; and (b) inhibits performance in a
hippocampal-dependent spatial T-maze task (Elmore, Burton et al. 2014). A single voxel was
placed over the left and right dorsal hippocampus. This location and a representative MR
spectrum are shown in Figure 3.5. Nine metabolites/metabolite combinations reached Cramer-
Rao lower bounds of SD < 20% in all subjects. These metabolites and their absolute
concentrations are shown in Table 3.3. Four of the metabolites were significantly reduced in the
PRRSV piglets including creatine/phosphocreatine (Cr+PCr), N-Acetylasparate (NAA), NAA/N-
Acetylaspartylglutamate (NAAG), and Myo-inositol (Ins).
3.5 Discussion
Respiratory infections during the neonatal period are common, yet little is known about
their impact on brain development. Progress in this area has been slow because studies in
human infants are either impossible, due to obvious ethical considerations, or extremely difficult.
Furthermore, results from rodent models commonly used to investigate neurodevelopment are
difficult to translate to human infants due to the substantial differences in brain development and
54
morphology. To overcome some of these barriers, the present study was conducted using a
highly tractable and translational piglet model.
Previously, we used MRI and a longitudinal study design and showed that brain growth
in the domestic piglet is similar in many regards to the human neonate (Conrad, Dilger et al.
2012). Logistical modeling suggested that total brain volume from birth to 4 wks of age
increased 45%. Therefore, in the present study inoculation with PRRSV and the subsequent
infection coincided with a period of rapid brain growth. PRRSV is an enveloped, positive-sense,
single-stranded RNA virus that belongs to the Arteriviridae family within the order Nidovirales
(Done, Paton et al. 1996). In young pigs PRRSV primarily infects and replicates in cells of the
monocyte/macrophage lineage (Duan, Nauwynck et al. 1997). Once infected, macrophages can
migrate to lymphatic tissue and enter circulation allowing for the spread of PRRSV (Shin and
Molitor 2002). Either by eliciting a pro-inflammatory response that activates immune-to-brain
signaling pathways or by entering the CNS, PRRSV activates microglia and causes
neuroinflammation. In a recent study, several pro-inflammatory cytokines were elevated in
serum 20 d after inoculation with PRRSV, and a number of pro-inflammatory genes, including
interferon-γ, TNFα, and IL-1β, were up regulated in several brain regions at the same time post
inoculation (Elmore, Burton et al. 2014). Furthermore, the percentage of activated microglia, as
indicated by expression of MHC class II, was markedly increased in piglets with PRRSV
infection (Elmore, Burton et al. 2014). Microglial activation was positively correlated with fever,
and negatively correlated with food motivation and learning and memory (Elmore, Burton et al.
2014). In the present study, piglets were febrile throughout the study and circulating TNFα was
elevated in samples collected after MRI (i.e., 21 d after inoculation). Thus, it is reasonable to
conclude that PRRSV infected piglets experienced a sustained neuroinflammatory response
throughout the study period.
We employed multiple MRI-based techniques to assess the effects of PRRSV infection
on piglet brain macrostructure (VBM), microstructure (DTI), and biochemistry (MRS). Although
55
this may be the first study to use a MRI-based approach to investigate if early-life infection
affects brain development, we recently used a similar strategy to explore brain development in
piglets born small-for-gestational-age (i.e., a model of intrauterine growth restriction; Radlowski
et al.) and piglets provided formula supplemented with phospholipids and gangliosides
(Radlowski et al.). In the present study, VBM revealed multiple brain areas in PRRSV piglets
with less gray matter volume. The two most significant clusters were located bilaterally in the
primary visual cortex. White matter volume was also reduced in these regions. The primary
visual cortex develops quickly with a high rate of synaptogenesis after birth and, therefore, may
be sensitive to environmental insults at this time (Johnson 1990; Bourgeois and Rakic 1993;
Gogtay, Giedd et al. 2004). Microglia are vital for synaptogenesis in the primary visual cortex
and activation may disrupt normal development (Rochefort, Quenech'du et al. 2002). While we
wish to be cautious not to overstate the significance of these findings in neonatal piglet model, it
is worth noting that a neuroimmune-based developmental hypothesis has been proposed for
several psychiatric diseases in humans that are associated with deficiencies in the primary
visual cortex (Meyer, Feldon et al. 2011). First, the primary visual cortex in autistic patients
appears normal, but other areas, such as the fusiform gyrus, change (van Kooten, Palmen et al.
2008). It is thought that the visual processing abnormalities in autism may be due to top-down
processing differences (Hadjikhani, Chabris et al. 2004). Second, some schizophrenia patients
present with visual processing problems which are thought to be due to a disruption in the early
development in the visual pathway (Butler, Schechter et al. 2001). Postmortem histological
analysis has shown that schizophrenia patients have roughly a 25% reduction in neuron number
and volume in the primary visual cortex compared to controls (Dorph-Petersen, Pierri et al.
2007). The reduced volumes seen in the PRRSV piglets maybe indicative of the sensitivity to
inflammation and may result in long term consequences. Either delays in synaptogenesis or
degeneration of established connections could explain the volume differences. Additional
56
studies to characterize the histological changes in the primary visual cortex of infected piglets
are warranted.
In addition to voxel-based morphometry, region of interest analysis was used to estimate
volumes of 19 brain regions. Changes were found in the ventricular system, including the
cerebral aqueduct, lateral, and fourth ventricle, where PRRSV piglets had an expanded volume.
Increased lateral ventricle volumes have been shown in both autism and schizophrenia (Chua
and McKenna 1995; Piven, Arndt et al. 1995; Palmen, Hulshoff Pol et al. 2005; Rosa,
Schaufelberger et al. 2010). We also found a sexual dimorphism in the corpus callosum where
males had a smaller corpus callosum than females, similar to humans (Ardekani, Figarsky et al.
2012).
Myelination begins in late gestation, has its highest rate of formation in early-life, and is
susceptible to inflammation (Nakagawa, Iwasaki et al. 1998; Lenroot and Giedd 2006; Abraham,
Vincze et al. 2010; Favrais, van de Looij et al. 2011). Disruptions to white matter tracts,
including the corpus callosum, have been found in psychiatric disease (Alexander, Lee et al.
2007; Cheung, Cheung et al. 2008; Cheung, Chua et al. 2009). Here, we found no differences
in the FA values across the whole brain and white matter. There was a trend in the corpus
callosum where the PRRSV piglets had decreased FA values which may indicate a disruption in
myelination. Previous studies in rodents have found that inflammatory cytokines block
oligodendrocytes maturation and cause axonopathy including reduced myelinated axon
diameter (Favrais, van de Looij et al. 2011). The inflammatory response that is seen in the
PRRSV piglets includes an increase in pro-inflammatory cytokines and this may be a
mechanism for the changes seen (Elmore, Burton et al. 2014). Additionally, we did not find any
differences in MD suggesting that there were no changes in cellular integrity. Further analysis is
needed in this model to determine if other white matter tracts are affected by the infection.
MR-spectroscopy is a powerful tool to quantify chemical metabolite concentrations non-
invasively (Soares and Law 2009). Nine metabolites were quantified in the hippocampus with
57
four being significantly lower in the PRRSV piglets including creatine/phosphocreatine, myo-
inositol, NAA, and NAAG. Reductions in creatine and myo-inositol suggest an energy
imbalance and decrease in glial cells or delayed gliogenesis compared to controls. NAA is a
marker of the density and viability of neurons and is reduced in many neuroinflammatory
conditions (Soares and Law 2009). The reduced NAA in the PRRSV piglets may be a sign of
neuronal loss or developmental impairment. Similar findings have been shown in hypoxia
ischemia models using piglets suggesting that energy-dysregulation, including secondary
energy failure, may be present after early-life infection (Munkeby, De Lange et al. 2008).
In conclusion, the present study found that PRRSV infection causes structural changes
in gray and white matter, as well as neurochemical changes. Due to the small number of
animals used for this study, we were limited in the analysis of covariates in VBM including sex.
Further histological analysis of the primary visual cortex and corpus callosum would be
beneficial to further characterize the changes seen. Regardless, early-life infection and
inflammation have a significant impact on normal brain development. In future studies, it will be
important to examine long-term endpoints to determine the lasting impact of early-life infection.
58
3.6 Figures and Tables
2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30
0
2
4
6
8
10
ControlPRRSV
A
Day of Age
Wei
gh
t (k
g)
7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 2837
38
39
40
41
* *# *
# #
#
## #
**
#
#*
# #
* = <0.05# = <0.01
B
Day of Age
Te
mp
era
ture
( C
)
7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 282.0
2.2
2.4
2.6
2.8
3.0
3.2
C
Day of Age
Fe
edin
g S
core
Figure 3.1. PRRSV piglets show a febrile response and reduced feeding score (A to C). There were no differences in weight between the two treatments (A). PRRSV piglets had significantly higher temperatures starting on the third day that lasted throughout the study. PRRSV piglets also had a significant reduction in feeding score. Data are shown as the mean ± SEM (*p<0.05, #p<0.01; n=6 per treatment).
59
Figure 3.2. VBM analysis showing clusters where PRRSV pigs had significantly less gray matter than controls. Axial slices are shown with a 1 mm slice thickness. The color bar indicates significance level (pseudo t-statistic).
60
Figure 3.3. VBM analysis showing areas where PRRSV pigs had significantly less white matter than controls. Axial slices are shown with a 1 mm slice thickness. The color bar indicates significance level (pseudo t-statistic).
61
0.32
0.34
0.36F
ract
ion
al A
nis
otr
op
y
0.26
0.28
0.30
0.32
#
Fra
ctio
nal
An
iso
tro
py
0.30
0.31
0.32
0.33
0.34ControlPRRSV
Fra
ctio
nal
An
iso
tro
py
G
A B C
0.0
0.5
1.0
1.5
Mea
n D
iffu
siv
ity
0.0
0.5
1.0
1.5
Mea
n D
iffu
siv
ity
0.0
0.5
1.0
1.5Control
PRRSV
Mea
n D
iffu
siv
ity
D E F
Figure 3.4. PRRSV piglets have a trend towards reduced FA in the corpus callosum (A to G). There were no differences in FA found in whole brain (A) and white matter (C). There was a trend that the controls had a higher FA than PRRSV in the corpus callosum (B; p=0.0637). There were no differences in MD in whole brain (D), corpus callosum (E), or white matter (F). A representative FA map in a axial, sagittal, and coronal slice from left to right is shown (G). Data are shown as the mean ± SEM (n=6 per treatment). Units for mean diffusivity are 10-3 mm2/s.
62
Figure 3.5. Representative proton MR spectroscopy spectrum and location of the voxel placement. The single voxel was placed across the left and right dorsal hippocampus for each animal.
63
Table 3.1. Brain region volume estimations. Volume averages for 22 regions are reported by sex and treatment. Significance due to sex, treatment, and the interaction are shown.
64
Cluster Cluster
level Local Maxima Coordinates*
Comparison Anatomic Region (voxels) (P-value) x y z Pseudo-t
Gray Matter
Left Primary Visual Ctx 616 .0032 -9 2 16 4.82
Control>PRRSV
PRRSV>Control
Right Primary Visual Ctx 530 .0022 10 1 17 4.44
Left Insular Ctx 118 .0032 -8 25 4 3.32
Prefrontal Ctx 169 .0043 0 22 -2 3.04
Left Auditory Ctx 30 .0087 -18 6 8 2.58
Right Fusiform Gyrus 22 .0087 15 -11 1 2.54
Left Fusiform Gyrus 175 .0054 -13 -10 3 2.51
Right Primary Visual Ctx 69 .0065 6 -3 20 2.39
No significant findings
White Matter
Control>PRRSV
PRRSV>Control
Left Primary Visual Ctx 208 .0022 -6 1 17 4.60
Right Primary Visual Ctx 615 .0032 8 -1 17 4.51
Right Primary Somatosensory Ctx 57 .0097 12 15 15 3.94
Left Primary Somatosensory Ctx 436 .0032 -11 9 17 3.61
.0032 -13 2 14 3.48
Left Primary Somatosensory Ctx 33 .0087 -8 22 15 2.35
Prefrontal Cortex 311 .0097 3 31 1 2.59
Right Internal Capsule 28 .0065 8 9 10 1.75
Left Hippocampus 50 .0097 -6 -6 2 1.73
Left Internal Capsule 59 .0097 -7 9 10 1.73
Table 3.2. Voxel-based morphometry analysis of gray and white matter differences. Significant clusters of at least 20 voxels are shown with the cluster size (voxels), cluster significance, location, and pseudo-t value.
65
Table 3.3. Absolute metabolite concentration in the hippocampus. Single voxel MRS for nine metabolites that met criteria for analysis (%S.D.<20). Data are shown as the mean ± SEM (n=6 per treatment).
Metabolite Control (SEM) PRRSV (SEM) P-value Creatine + Phosphocreatine (Cr + PCr) 4.3395 (0.0915) 3.6992 (0.1163) 0.0015 N-Acetylaspartate (NAA) 5.3473 (0.0579) 4.6953 (0.1977) 0.0101 NAA + N-Acetylaspartylglutamate (NAAG) 5.7728 (0.0286) 5.0563 (0.1845) 0.0033 Myoinositol (Ins) 9.5028 (0.4302) 8.2737 (0.3395) 0.0488 Phosphocholine + Glycerophosphocholine (GPC + PCh) 1.6775 (0.0506) 1.6253 (0.0561) 0.5056 Glutamate + Glutamine (Glu + Gln) 6.6178 (0.3099) 7.4558 (0.6084) 0.2478 MM20 8.4418 (0.5506) 9.0770 (0.4523) 0.3937 MM09+Lip09 4.4567 (0.0708) 4.5043 (0.0941) 0.6941 MM20+Lip20 9.2922 (0.4385) 9.7607 (0.4048) 0.4506
66
3.7 References
Abraham, H., A. Vincze, et al. (2010). "Myelination in the human hippocampal formation from midgestation to adulthood." Int J Dev Neurosci 28(5): 401-410.
Alexander, A. L., J. E. Lee, et al. (2007). "Diffusion tensor imaging of the corpus callosum in
Autism." NeuroImage 34(1): 61-73. Ardekani, B. A., K. Figarsky, et al. (2012). "Sexual Dimorphism in the Human Corpus Callosum:
An MRI Study Using the OASIS Brain Database." Cerebral Cortex. Ashburner, J. (2007). "A fast diffeomorphic image registration algorithm." NeuroImage 38(1): 95-
113. Bourgeois, J. and P. Rakic (1993). "Changes of synaptic density in the primary visual cortex of
the macaque monkey from fetal to adult stage." The Journal of Neuroscience 13(7): 2801-2820.
Butler, P. D., I. Schechter, et al. (2001). "Dysfunction of early-stage visual processing in
schizophrenia." Am J Psychiatry 158(7): 1126-1133. Cheung, C., S. E. Chua, et al. (2009). "White matter fractional anisotrophy differences and
correlates of diagnostic symptoms in autism." J Child Psychol Psychiatry 50(9): 1102-1112.
Cheung, V., C. Cheung, et al. (2008). "A diffusion tensor imaging study of structural
dysconnectivity in never-medicated, first-episode schizophrenia." Psychol Med 38(6): 877-885.
Chua, S. E. and P. J. McKenna (1995). "Schizophrenia--a brain disease? A critical review of
structural and functional cerebral abnormality in the disorder." Br J Psychiatry 166(5): 563-582.
Conrad, M. S., R. N. Dilger, et al. (2012). "Brain Growth of the Domestic Pig (Sus scrofa) from 2
to 24 Weeks of Age: A Longitudinal MRI Study." Dev Neurosci 34(4): 291-298. Conrad, M. S., R. N. Dilger, et al. (2012). "Magnetic resonance imaging of the neonatal piglet
brain." Pediatr Res 71(2): 179-184. Dantzer, R. and K. W. Kelley (2007). "Twenty years of research on cytokine-induced sickness
behavior." Brain, Behavior, and Immunity 21(2): 153-160. Dawson, H. D. (2011). A Comparative Assessment of the Pig, Mouse and Human Genomes.
The Minipig in Biomedical Research, CRC Press: 323-342. Dickerson, J. W. T. and J. Dobbing (1967). "Prenatal and postnatal growth and development of
the central nervous system of the pig." Proc Biol Sci 166(1005): 384-395.
67
Dietrich, R. B., W. G. Bradley, et al. (1988). "MR evaluation of early myelination patterns in normal and developmentally delayed infants." American Journal of Roentgenology 150(4): 889-896.
Dilger, R. N. and R. W. Johnson (2010). "Behavioral assessment of cognitive function using a
translational neonatal piglet model." Brain Behav Immun 24(7): 1156-1165. Dobbing, J. and J. Sands (1979). "Comparative aspects of the brain growth spurt." Early Hum
Dev 3(1): 79-83. Done, S. H., D. J. Paton, et al. (1996). "Porcine reproductive and respiratory syndrome (PRRS):
a review, with emphasis on pathological, virological and diagnostic aspects." Br Vet J 152(2): 153-174.
Dorph-Petersen, K.-A., J. N. Pierri, et al. (2007). "Primary visual cortex volume and total neuron
number are reduced in schizophrenia." The Journal of Comparative Neurology 501(2): 290-301.
Duan, X., H. J. Nauwynck, et al. (1997). "Virus quantification and identification of cellular targets
in the lungs and lymphoid tissues of pigs at different time intervals after inoculation with porcine reproductive and respiratory syndrome virus (PRRSV)." Vet Microbiol 56(1-2): 9-19.
Elmore, M. R., M. D. Burton, et al. (2014). "Respiratory viral infection in neonatal piglets causes
marked microglia activation in the hippocampus and deficits in spatial learning." J Neurosci 34(6): 2120-2129.
Fang, M., D. E. Lorke, et al. (2005). "Postnatal changes in functional activities of the pig’s brain:
a combined functional magnetic resonance imaging and immunohistochemical study." Neurosignals 14(5): 222-233.
Favrais, G., Y. van de Looij, et al. (2011). "Systemic inflammation disrupts the developmental
program of white matter." Ann Neurol 70(4): 550-565. Felix, B., M. E. Leger, et al. (1999). "Stereotaxic atlas of the pig brain." Brain Res Bull 49(1-2):
1-137. Gogtay, N., J. N. Giedd, et al. (2004). "Dynamic mapping of human cortical development during
childhood through early adulthood." Proc Natl Acad Sci U S A 101(21): 8174-8179. Hadjikhani, N., C. F. Chabris, et al. (2004). "Early visual cortex organization in autism: an fMRI
study." Neuroreport 15(2): 267-270. Hall, C. B., G. A. Weinberg, et al. (2009). "The burden of respiratory syncytial virus infection in
young children." N Engl J Med 360(6): 588-598. Huttenlocher, P. R. (1979). "Synaptic density in human frontal cortex - developmental changes
and effects of aging." Brain Res 163(2): 195-205. Jakobsen, S., K. Pedersen, et al. (2006). "Detection of α2-Adrenergic Receptors in Brain of
Living Pig with 11C-Yohimbine." J Nucl Med 47(12): 2008-2015.
68
Jenkinson, M., C. F. Beckmann, et al. (2012). "FSL." NeuroImage 62(2): 782-790. Johnson, M. H. (1990). "Cortical maturation and the development of visual attention in early
infancy." J. Cognitive Neuroscience 2(2): 81-95. Lenroot, R. K. and J. N. Giedd (2006). "Brain development in children and adolescents: insights
from anatomical magnetic resonance imaging." Neurosci Biobehav Rev 30(6): 718-729. Meurens, F., A. Summerfield, et al. (2012). "The pig: a model for human infectious diseases."
Trends in microbiology 20(1): 50-57. Meyer, U., J. Feldon, et al. (2011). "Schizophrenia and autism: both shared and disorder-
specific pathogenesis via perinatal inflammation?" Pediatr Res 69(5 Pt 2): 26R-33R. Munkeby, B. H., C. De Lange, et al. (2008). "A piglet model for detection of hypoxic-ischemic
brain injury with magnetic resonance imaging." Acta Radiol 49(9): 1049-1057. Nakagawa, H., S. Iwasaki, et al. (1998). "Normal myelination of anatomic nerve fiber bundles:
MR analysis." American Journal of Neuroradiology 19(6): 1129-1136. Nichols, T. E. and A. P. Holmes (2002). "Nonparametric permutation tests for functional
neuroimaging: a primer with examples." Hum Brain Mapp 15(1): 1-25. Palmen, S. J., H. E. Hulshoff Pol, et al. (2005). "Increased gray-matter volume in medication-
naive high-functioning children with autism spectrum disorder." Psychol Med 35(4): 561-570.
Piven, J., S. Arndt, et al. (1995). "An MRI study of brain size in autism." Am J Psychiatry 152(8):
1145-1149. Provencher, S. (September 14, 2013). "LCModel and LCMgui User's Manual." from
http://www.s-provencher.com/pub/LCModel/manual/manual.pdf. Rice, D. and S. Barone, Jr. (2000). "Critical periods of vulnerability for the developing nervous
system: evidence from humans and animal models." Environ Health Perspect 108 Suppl 3: 511-533.
Rochefort, N., N. Quenech'du, et al. (2002). "Microglia and astrocytes may participate in the
shaping of visual callosal projections during postnatal development." J Physiol Paris 96(3-4): 183-192.
Rosa, P. G., M. S. Schaufelberger, et al. (2010). "Lateral ventricle differences between first-
episode schizophrenia and first-episode psychotic bipolar disorder: A population-based morphometric MRI study." World J Biol Psychiatry 11(7): 873-887.
Saikali, S., P. Meurice, et al. (2010). "A three-dimensional digital segmented and deformable
brain atlas of the domestic pig." J Neurosci Meth 192(1): 102-109. Sawada, M., Y. Itoh, et al. (1993). "Expression of cytokine receptors in cultured neuronal and
glial cells." Neurosci Lett 160(2): 131-134.
69
Shin, J. H. and T. W. Molitor (2002). "Localization of porcine reproductive and respiratory syndrome virus infection in boars by in situ riboprobe hybridization." J Vet Sci 3(2): 87-96.
Soares, D. P. and M. Law (2009). "Magnetic resonance spectroscopy of the brain: review of
metabolites and clinical applications." Clinical Radiology 64(1): 12-21. Thibault, K. L. and S. S. Margulies (1998). "Age-dependent material properties of the porcine
cerebrum: effect on pediatric inertial head injury criteria." J Biomech 31(12): 1119-1126. van Kooten, I. A., S. J. Palmen, et al. (2008). "Neurons in the fusiform gyrus are fewer and
smaller in autism." Brain 131(Pt 4): 987-999.
70
CHAPTER 4
EARLY POSTNATAL RESPIRATORY VIRAL INFECTION ALTERS HIPPOCAMPAL NEUROGENESIS, CELL FATE, AND NEURON MORPHOLOGY IN THE NEONATAL PIGLET 4.1 Abstract
Respiratory viral infections are common during the neonatal period in humans, but little
is known about how early-life infection impacts brain development. The current study used a
neonatal piglet model with porcine reproductive and respiratory syndrome virus (PRRSV) to
evaluate how chronic neuroinflammation affects hippocampal neurogenesis and neuron
morphology. Piglets in the neurogenesis study received one bromodeoxyuridine injection on
postnatal day (PD) 7 and then were inoculated with PRRSV. Piglets were sacrificed at PD 28
and the number of number of BrdU+ cells and cell fate were quantified in the dentate gyrus.
PRRSV piglets showed a 24% reduction in the number of newly divided cells forming neurons.
Approximately 15% of newly divided cells formed microglia, but this was not affect by sex or
PRRSV. Additionally, there was a sexual dimorphism of new cell survival in the dentate gyrus
where males had more cells than females, and PRRSV infection caused a decreased survival in
males only. Golgi impregnation was used to characterize dentate granule cell morphology.
Sholl analysis revealed that PRRSV causes a change in inner granule cell morphology where
the first branch point is extended further away from the cell body. Males had more complex
dendritic arbors than females in the outer granule cell layer, but this was not affected by
PRRSV. There were no changes to dendritic spine density or morphology distribution. These
findings suggest that early-life infection can impact brain development.
71
4.2 Introduction
Respiratory infections are common in neonates, but little is known about their impact on
short- and long-term brain development (Hall, Weinberg et al. 2009). This is important,
however, because brain development in the neonatal period is characterized by extensive
dendritic growth, synaptogenesis, gliogenesis, and myelination (Huttenlocher 1979; Dietrich,
Bradley et al. 1988; Rice and Barone 2000). Furthermore, although neurogenesis occurs
primarily during the prenatal period, the subependyma of the lateral ventricles and granule cell
layer of the hippocampal dentate gyrus are two regions where neurogenesis continues postnatal
and into adulthood (Ekdahl 2012). During infection, immune-to-brain signaling pathways
activate brain microglia, causing increased production of pro-inflammatory cytokines (Dantzer
and Kelley 2007). Developing and mature neurons and glia have numerous pro-inflammatory
cytokine receptors and are sensitive to inflammatory conditions (Dantzer and Kelley 2007).
Therefore, understanding the impact of early-life infection on brain development is crucial. A
number of psychiatric disorders are associated with neuroimmune alterations and are thought to
have developmental origins (Boksa 2010; Meyer, Feldon et al. 2011).
Studies in adult animals with a fully mature brain suggest infection-related
neuroinflammation inhibits neurogenesis and alters neuron morphology. One brain area that is
especially vulnerable to inflammatory insults and that is important for spatial learning and
memory is the hippocampus (Elmore, Dilger et al. 2012). Peripheral immune activation with
bacterial lipopolysaccharide (LPS) increased the expression of pro-inflammatory cytokines like
interleukin (IL)-1β, IL-6, and tumor necrosis factor-alpha (TNF-α) in the brain (Kelley, Bluthé et
al. 2003) and inhibited the survival of newborn neurons in the dentate gyrus without impacting
cell proliferation (Ekdahl, Claasen et al. 2003). In vitro and in vivo studies also showed that pro-
inflammatory cytokines affect neural precursor generation, differentiation, and survival
(Vallieres, Campbell et al. 2002; Cacci, Ajmone-Cat et al. 2008; Wu, Hein et al. 2012; Wu,
Montgomery et al. 2013).
72
Pro-inflammatory cytokines can further impair synaptic plasticity by inhibiting production
of neurotrophins like BDNF, inhibiting long-term potentiation, and altering the architecture of
dendrites (Lynch 2002; Milatovic, Zaja-Milatovic et al. 2003; Richwine, Parkin et al. 2008; Tong,
Balazs et al. 2008; Jurgens, Amancherla et al. 2012). Although these and many other studies
show that infection and neuroinflammation inhibits neurogenesis and alters neuron morphology
in the adult brain, there has been little research on how infection affects neurogenesis or neuron
morphology in the critical early postnatal period (Green and Nolan 2014).
Therefore, the goal of this study was to determine the impact of respiratory viral infection
in the early postnatal period on hippocampal neurogenesis, cell fate, and neuron morphology in
a domestic piglet model. The piglet is an ideal model for this type of investigation because it
has a gyrencephalic brain that grows and develops similar to human infants. (Conrad, Dilger et
al. 2012). In the present study, piglets were experimentally infected on postnatal day (PD) 7
with porcine reproductive and respiratory syndrome virus (PRRSV) and neurogenesis and
neuron morphology were determined with brain tissue collected PD 28. PRRSV activates
microglia in the hippocampus of piglets and causes increased pro-inflammatory cytokine
production and deficits in hippocampal-dependent learning and memory (Elmore, Burton et al.
2014). Here we show that PRRSV infection impacts new cell survival, cell fate, and granule cell
morphology. These findings are the first to show that a live viral infection in the neonatal period
alters neurogenesis and neuron morphology.
4.3 Materials and Methods
Animals, Housing, and Feeding
Naturally farrowed crossbred piglets from six separate litters (20 males and 20 females)
were obtained from the University of Illinois swine herd. Piglets were brought to the biomedical
animal facility on PD 2 (to allow for colostrum consumption from the sow) and placed in
individual cages (0.76 m L x 0.58 m W x 0.47 m H) designed for neonatal piglets (Elmore,
73
Burton et al. 2014). Each cage was positioned in a rack, with stainless steel perforated side
walls and clear acrylic front and rear doors within one of two separate but identical disease
containment chambers that have been described (Elmore, Burton et al. 2014). Each cage was
fitted with flooring designed for neonatal animals (Tenderfoot/NSR, Tandem Products, Inc.,
Minneapolis, MN, USA). A toy (plastic Jingle BallTM, Bio-Serv, Frenchtown, NJ, USA) was
provided to each piglet. Room temperature was maintained at 27ºC and each cage was
equipped with an electric heat pad (K&H Lectro-KennelTM Heat Pad, K&H Manufacturing, LLC,
Colorado Springs, CO, USA). Piglets were maintained on a 12-h light/dark cycle; however,
during the dark cycle minimal lighting was provided.
Piglets were fed a commercial sow milk replacer (Advance Liqui-Wean, Milk Specialties
Co., Dundee, IL, USA). Milk was reconstituted daily to a final concentration of 206 g/L using tap
water and supplied at a rate of 285 mL/kg BW (based on daily recorded weights) to a stainless
steel bowl via a peristaltic pump (Control Company, Friendswood, TX). Using this automated
feeding system (similar to that described previously (Dilger and Johnson 2010)), piglets
received their daily allotted milk over 18 meals (once per hour). All animal experiments were in
accordance with the National Institute of Health Guidelines for the Care and Use of Laboratory
Animals and approved by the University of Illinois at Urbana-Champaign Institutional Animal
Care and Use Committee.
Experimental Design and Treatments
Upon arrival, piglets were assigned to either the control group or the PRRSV infection
group based on sex, litter of origin, and body weight. To determine the effect of early-life viral
infection on neurogenesis and cell fate, on PD 7, 24 piglets (12 males and 12 females) were
injected i.p. with BrdU (50 mg/kg, Sigma, St. Louis, MO, USA) and then inoculated intranasal
with either 1mL of 1x10^5 50% tissue culture infected dose (TCID 50) of live PRRSV (strain
74
P129-BV, obtained from the School of Veterinary Medicine at Purdue University, West
Lafayette, IN, USA) or sterile phosphate buffered saline (PBS). Two PRRSV piglets developed
diarrhea and were unable to complete the study. All remaining piglets (control male, n=6;
control female, n=6; PRRSV male, n=5; PRRSV female, n=5) were sacrificed at PD 28. To
determine the effect of early-life viral infection on hippocampal neuron morphology, a total of 16
piglets (8 males and 8 females) piglets were inoculated with PRRSV or sterile PBS at PD 7.
One PRRSV piglet developed diarrhea and was unable to complete the study. All remaining
piglets were sacrificed at PD 31 (control male, n=4; control, female n=4; PRRSV male, n=4;
PRRSV female, n=3).
Assessment of Infection
Daily body weights (kg) were obtained to monitor piglets’ growth. In addition, daily rectal
temperatures were obtained starting at PD 7. The willingness of the piglets to consume their
first daily meal was determined starting at PD 7 using a feeding score (1 = no attempt to
consume the milk; 2 = attempted to consume the milk, but did not finish within 1 min; 3 =
consumed all of the milk within 1 min).
The presence of PRRSV antibodies in the serum of all piglets at the end of the study
was analyzed by the Veterinary Diagnostic Laboratory (University of Illinois, Urbana, Illinois)
using a PRRSV-specific ELISA kit (IDEXX Laboratories, Westbrook, ME, USA). This assay has
98.8% sensitivity and 99.9% specificity, with an S/P ratio of >0.4 indicating a positive sample.
Serum TNF-α levels at the end of the study were determined using porcine-specific sandwich
enzyme immunoassays (R&D Systems, Minneapolis, MN, USA).
75
Perfusions and Tissue Processing
For the neurogenesis and cell fate study, all animals were sacrificed and perfused at PD
28. A telazol:ketamine:xylazine solution was administered intramuscularly at 4.4 mg/kg BW
(50.0 mg of tiletamine, plus 50.0 mg of zolazepam, reconstituted with 2.50 mL ketamine [100
g/L] and 2.50 mL xylazine [100 g/L]; Fort Dodge Animal Health, Fort Dodge, IA, USA). Eyeblink
and pain reflexes were tested to confirm deep anesthesia before piglets were perfused
transcardially with a phosphate buffer solution (PBS) and a 4% paraformaldehyde/PBS solution.
Brains were extracted and postfixed overnight. The hippocampus was removed and transferred
to PBS with 30% sucrose until the tissue sank (~2 days). The entire hippocampus was
sectioned using a cryostat into 40 µm sections in an axial plane from dorsal to ventral and
collected into a 12-well plate. Six sections were collected into each well with a distance of 480
µm separating each well. Tissues from dorsal region of the hippocampus were used for
staining.
BrdU-DAB
The DAB staining was adapted from previously described work (Kohman, DeYoung et
al. 2012). Briefly, free floating sections were washed in Tris-buffering solution (TBS) and
treated with 0.6% hydrogen peroxide solution for 30 min. Next, sections were placed in 50%
de-ionized formamide for 90 min to denature DNA. Sections were then placed in a 10% 20x
saline sodium citrate buffer for 15 min, 2N hydrochloric acid for 30 min at 37°C, and then 0.1 M
boric acid (pH 8.5) for 10 min. After rinsing, sections were blocked with a solution of 0.3%
Triton-X and 3% goat serum in TBS (TBS-X) for 30 min. The sections were then incubated with
the primary antibody rat anti-BrdU (1:200; AbD Serotec, Raleigh, NC, USA) in TBS-X for 72 h at
4°C. Sections were then rinsed with TBS, blocked with TBS-X for 30 min, and then incubated
76
with a biotinylated goat anti-rat secondary antibody (1:250, Jackson ImmunoResearch
Laboratories, West Grove, PA, USA) for 100 min. The ABC system (Vector, Burlingame, CA,
USA) and diaminobenzidine kit (DAB; Sigma, St. Louis, MO, USA) were used for the
chromogen.
Immunofluorescence
For the BrdU/NeuN double staining and the BrdU/GFAP/IBA-1 triple staining, a similar
procedure was used as with DAB staining. The primary antibodies consisted of rat anti-BrdU
(1:100; AbD Serotec, Raleigh, NC, USA), mouse anti-NeuN (1:50, Millipore, Temecula, CA,
USA), mouse anti-GFAP (1:50, Santa Cruz Biotechnology, Santa Cruz, CA, USA), and rabbit
anti-IBA1 (1:1000, Wako Chemicals, Richmond, VA, USA). All secondary fluorescent
antibodies (All 1:250, Alexa-488, Cy3, and Alexa-647) were made in goat and incubated for 3 h
at room temperature.
BrdU-DAB Image Analysis
Slides containing the DAB-stained tissue were digitized using a Nanozoomer digital
pathology system at 20x lens power (Hamamatsu Photonics, Hamamatsu, Japan). The
hippocampus of each tissue section was then exported at 10x resolution. Using ImageJ,
regions of interest (ROI) were drawn for the suprapyramidal blade, infrapyramidal blade, and
hilus of the dentate gyrus and ROI volumes measured. For the suprapyramidal and
infrapyramidal blade, the granule cell layer and subgranular zone were included. An overview
of the hippocampus with a highlighted view of the suprapyramidal blade can be seen in Figure
4.1. ImageJ was used to automatically count the number of positively labeled cells in all three
77
ROIs. The data are expressed as BrdU-positive cells per cubic micrometer. Thirty-two sections
(~2400 positive cells) were used to validate the automated methods versus manual hand
counting. Correlation analysis validated the automatic method with a slope no different from a
1-to-1 line and a Pearson r value of 0.99. Unbiased estimation was used to correct for cells that
could be intersecting with either the top or bottom of the tissue section. An average BrdU-
positive nucleus was 6.5 μm in diameter (320 sampled) which is 16.25% of 40 μm BrdU-positive
cell counts were multiplied by 0.8375 for unbiased estimation correction.
Immunofluorescence Analysis
A Zeiss LSM 700 confocal microscope (20x objective) was used to acquire z-stack
images with a 0.5 micron slice thickness in the dentate gyrus including the granule cell layer and
subgranular zone. Images were deconvoluted using Autoquant (Media Cybernetics, Rockville,
MD, USA). For the BrdU/NeuN co-localization, cells from both the suprapyramidal and
infrapyramidal blades were acquired but later combined for analysis due to no differences
between the two regions. Only cells from the suprapyramidal blade were sampled for the
BrdU/GFAP/IBA-1 triple staining.
Raters were blinded to the treatments and manually counted the BrdU-positive and the
number of either NeuN, GFAP, or IBA-1-positive cells that co-localized with BrdU-positive
nuclei. Data is expressed as the proportion of BrdU-positive cells that co-localized with another
cell marker.
78
Hippocampal Neuronal Architecture Staining
To determine the effect of early-life viral infection on hippocampal neuron architecture,
animals were euthanized at PD 31. A similar dose of TKX (see above) was used for induction
of anesthesia and the animals were euthanized by intracardiac injection of sodium pentobarbital
(86.0 mg/kg of B.W.; Fatal Plus®, Vortech Pharmaceuticals, Dearborn, MI, USA). The brain
was removed and the left hippocampus was extracted and processed for Golgi-Cox staining as
previously described (Richwine, Parkin et al. 2008; Jurgens, Amancherla et al. 2012). Briefly,
the left hippocampus was submerged in Golgi-Cox solution for two weeks at which point daily
test slices were taken to track neuron filling. Tissues were removed after 4-wks, dehydrated,
and embedded in 12% celloiden. The dorsal hippocampus was sectioned at 140 µm and
mounted on glass slides. Experimenters responsible for neuron tracing and spine density
measurement were blinded to the conditions.
Neuron Selection and tracing
Hippocampal neuron morphology was quantified using a Zeiss Axio Imager A.1
microscope and a computer-based system (Neurolucida; MBF Bioscience, Williston, VT, USA).
NeuroExplorer (MBF Bioscience) was used for visualization and analysis of neuron tracings. An
example of a stained dentate granule cell neuron and tracing is shown in Figure 4.2. Dentate
granule cell neurons were selected from the suprapyramidal blade and were distinct from other
neurons, not truncated, and were well filled. The granule cell neurons were traced at 40X
magnification and then segments of the dendrites were captured at 100X magnification for
dendritic spine analysis. Previous research has shown that the complexity of granule cells in
the inner 2/3 of the granule cell layer (towards hilus) are different than cells in the outer 1/3
(towards molecular layer) (Green and Juraska 1985). Therefore, 5 granule cell neurons from
79
each region were traced and analyzed per pig. After tracing, an estimation of dendritic
complexity was determined by calculating the total dendritic length and intersections. Dendritic
tree morphology was analyzed using Sholl ring analysis. For the Sholl analysis, 3D concentric
spheres with an increasing radius (20 μm increments) were placed around the cell body. The
number of intersections of the dendrites and the concentric rings per radial distance from the
soma were quantified.
Quantification of Spine Density and Morphology
Spine density measurements were conducted on the same cells quantified for Sholl
analysis. For each dentate gyrus granule cell, three dendritic segments were traced. Only 2-5°
order branches and dendrites that were 20 µm or greater in length were selected. Each
segment was at least 50 µm away from the cell soma. After neuron tracing was completed,
dendritic spines were counted using Neurolucida. Spines were counted on both sides of the
dendritic segment and classified according to their shapes; either thin, stubby, mushroom,
filopodium, or branched (Tashiro and Yuste 2003). The density of spines is expressed as
number of spines per micron of dendrite.
Statistical Analysis
Data analysis was conducted using the MIXED procedure of SAS (SAS Institute Inc,
Cary, NC, USA). Sickness measures were analyzed as a two-way (treatment × day) repeated
measures ANOVA with trial as a blocking factor. There were no significant differences due to
sex, therefore it was not included in the final analysis of sickness symptoms. Neurogenesis
data were analyzed as a two-way (treatment × sex) ANOVA. Total dendritic length and total
80
intersections in the inner and outer layer were analyzed separately as a two-way (treatment ×
sex) ANOVA. Data from the Sholl analysis were analyzed as a three-way (treatment × sex ×
distance) repeated measures ANOVA with pig as a random effect and nested within treatment
and sex. When a main effect or interaction was significant, post hoc Student’s t tests using
Fisher’s least significant differences were used. Statistical significance was set at p < 0.05.
Data are presented as means ± SEM.
4.4 Results
PRRSV Infection and Measures of Sickness
Serum ELISA results indicated all control piglets were negative for PRRSV at the end of
the study, whereas all piglets inoculated with PRRSV were positive for PRRSV. Body weight,
rectal temperature, and the willingness to consume the first daily meal (feeding score) were
determined to provide an indication of the sickness response of piglets infected with PRRSV
(Figure 4.3). Analysis of body weight data showed a significant effect of day (F (30, 962) =
61.98, p < 0.001) and a day × treatment interaction (F (30, 962) = 2.25, p = 0.001), where
controls had higher weight gain toward the conclusion of the study. Overall, the effect of
treatment on body weight was not significant (F (1, 34) = 1.21). Analysis of feeding score data
revealed a significant effect of treatment (F (1, 34) = 34.67, p < 0.001), day (F (24, 734) = 1.99,
p = 0.003), and a treatment × day interaction (F (23, 734) = 2.08, p = .002), indicating PRRSV
piglets’ motivation to consume the first meal of the day was reduced. Analysis of rectal
temperature data showed a significant effect of treatment (F (1, 34) = 80.15), day (F (24, 766) =
7.34), and a treatment × day interaction (F (24, 766) = 3.49 (all, p < 0.001). PRRSV piglets
became febrile 3 d after inoculation and remained so during most of the experimental period. At
the conclusion of the study, plasma cytokine TNF-α concentration was significantly higher in
PRRSV piglets (209.2 ± 36.3 pg/mL) compared to controls (21.5 ± 4.3 pg/mL) (F (1, 34) = 32.97
81
p < 0.001). Collectively, these data indicate infection with PRRSV in the neonatal period
induced clinical signs of illness that persisted throughout the study period.
Hippocampal Cell Proliferation and Survival
BrdU was injected at PD 7, just before PRRSV inoculation, and brain tissue was
collected 3-wks later. Therefore, the number of BrdU+ cells represents the basal level of cell
division at PD 7, subsequent division of cells labeled at PD 7, and the effects of viral infection on
the survival of these labeled cells (Figure 4.4). Analysis of BrdU+ cell density in the
suprapyramidal blade revealed a significant effect of sex (F (1, 18) = 8.32, p = 0.01) and a sex ×
treatment interaction (F (1, 18) = 5.50, p = 0.03), whereby control males had higher density of
BrdU+ cells than control females (24,274 ± 2604 and 11,612 ± 2063 cells/mm3, respectively, p =
0.001) and PRRSV caused a reduction of BrdU+ cells in males only (14,445 ± 1582 cells/mm3,
p = 0.01). In the infrapyramidal blade, only the effect of sex was significant (F (1, 17) = 7.68, p
= 0.01). The density of BrdU+ cells in the hilus was also quantified. A significant effect of sex
(F (1, 18) = 8.04, p = 0.01) and a sex × treatment interaction (F (1, 18) = 6.8, p = 0.02) was
found, whereby female controls had a lower density of BrdU+ cells than male controls (3,624 ±
678 and 10,691 ± 1786 cells/mm3, respectively, p = 0.001) and whereas PRRSV numerically
reduced the number of BrdU+ cells in males, it numerically increased the number of BrdU+ cells
in females.
Cell Fate
Immunofluorescence was used to determine the fate of BrdU+ cells in the granule cell
layer and subgranular zone (Figure 4.5). To estimate the percentage of Brdu+ cells that
82
developed into mature neurons, the number of BrdU+ cells that co-labeled with NeuN was
determined. Two-way ANOVA of the percentage of double-labeled cells revealed a significant
effect of treatment (F (1, 18) = 34.42 p < 0.001) where in control piglets more than 80% of the
cells labeled with BrdU at PD 7 differentiated into mature neurons by PD 28 but in PRRSV
piglets only 57% of the cells labeled with BrdU at PD 7 differentiated into mature neurons. The
marked reduction in neurogenesis caused by PRRSV was similar in both males and females.
To estimate the percentage of BrdU+ cells that developed into microglia, the number of BrdU+
cells that co-labeled with IBA-1 was determined. Roughly 15% of the cells labeled with BrdU at
PD 7 differentiated into microglia by PD 28 and this was not influenced by sex (F (1, 18) = 1.05),
treatment (F (1, 18) = 0.15), or the sex × treatment interaction (F (1, 18) = 0.21). Tissue
sections were also stained with GFAP in an attempt to quantify the number of BrdU+ cells that
developed into astrocytes. However, due to the high density of astrocytes within the region
analyzed and problems staining two intracellular markers, we were unable to clearly identify
double labeled cells (data not shown).
Dendritic Arborization
Total dendritic length and total intersection in the Sholl analysis were used to quantify
the overall complexity of the inner and outer granule cell neurons (Figure 4.6). A significant
difference between the inner and outer granule cell neurons was evident (F (1, 22) = 9.86, p =
0.0048), therefore the two regions were analyzed separately for the overall complexity. Two-
way ANOVA of the total dendritic length and the number of intersections of the inner dentate
granule cell neurons showed neither an effect of sex, treatment or a sex × treatment interaction
(all p > 0.05). However, there was a significant effect of sex on dendritic length (F (1, 11) =
4.95, p = 0.048) and number of intersections (F (1, 11) = 5.99, p = 0.0328) for neurons in the
83
outer layer of the dentate gyrus where males had longer more complex dendrites than females.
A similar trend was found when analyzing dendritic length and intersections within each Sholl
interval. A repeated-measures ANOVA showed a significant effect of sex (F (1, 11) = 5.95, p =
0.0328) and distance from the soma (F (16, 176) = 55.02, p < 0.0001) on intersections for the
outer dentate granule cell neurons. Males (2.11 ± 0.11) had more intersections than females
(1.72 ± 0.12). Interestingly, analysis of the inner dentate granule cell neurons revealed a
significant effect of distance (17, 187) = 25.29, p < 0.0001) and a distance × treatment
interaction (F (17, 187) = 2.26, p = 0.0041). The altered dendritic arborization was due to
shifting of the initial branching points away from the cell soma (within 40-140 μm) in the PRRSV
piglets (p < 0.05).
Spine Density
For the inner and outer portion of the granule cell layer spines were counted on both
sides of the dendritic segment and classified according to their shape: thin, stubby, mushroom,
filopodium, or branched (Figure 4.7 and 4.8) Neither spine density or classification were affected
in either the inner or outer granule cell layers by sex, treatment, or the sex × treatment
interaction (all p > 0.05).
4.5 Discussion
Respiratory infections during early-life are common, but knowledge of their impact on
brain development is lacking. Due to ethical considerations and the difficulty of conducting
experiments in human neonates, progress in this area has been slow. Furthermore, the
translation of data from rodent neurodevelopmental models to human infants is difficult due to
84
substantial differences in brain morphology and development. To overcome several obstacles,
the current study was conducted using a highly tractable and translational piglet model.
Previously, we have characterized the normal brain growth of the domestic pig using
magnetic resonance imaging (Conrad, Dilger et al. 2012). This study showed that piglet brain
growth is similar to human neonates and that from birth to 4-wks of age, total brain volume
increased 45% (Conrad, Dilger et al. 2012). Therefore, in the present study inoculation with
PRRSV and subsequent infection occurred in a period of substantial brain growth. PRRSV is
an enveloped, positive-sense, single-stranded RNA virus that belongs to the Arteriviridae family
within the order Nidovirales (Done, Paton et al. 1996). In young pigs PRRSV primarily infects
and replicates in cells of the monocyte/macrophage lineage (Duan, Nauwynck et al. 1997).
Once infected, macrophages can migrate to lymphatic tissue and enter circulation allowing for
the spread of PRRSV (Shin and Molitor 2002). PRRSV can activate microglia and cause
production of pro-inflammatory cytokines in the brain either through immune-to-brain signaling
pathways or by entering the CNS. In a recent study, several pro-inflammatory cytokines were
elevated in serum 20 d after inoculation with PRRSV, and a number of pro-inflammatory genes,
including interferon-γ, TNF-α, and IL-1β, were up regulated in several brain regions at the same
time post inoculation (Elmore, Burton et al. 2014). Furthermore, the percentage of activated
microglia, as indicated by expression of MHC class II, was markedly increased in piglets with
PRRSV infection (Elmore, Burton et al. 2014). Microglial activation was positively correlated
with fever, and negatively correlated with food motivation and learning and memory (Elmore,
Burton et al. 2014). In the present study, piglets exhibited a sustained febrile response and
increased circulating TNF-α at the conclusion of the experiment. This strongly suggests that
PRRSV piglets had a sustained neuroinflammatory response throughout the study period.
Both in vivo and in vitro studies have shown that inflammation can lead to altered cell
fate for newly divided cells. IL-1β causes a switch from neuronal to glial differentiation in vitro
85
(Green, Treacy et al. 2012). IL-1β can also inhibit the proliferation of neural progenitor cells and
proliferation of newly born neurons (Green and Nolan 2014). Similar to the in vitro data,
prenatal and early postnatal immune activation in rodents causes a reduction in neuronal
differentiation (Bland, Beckley et al. 2010; Graciarena, Depino et al. 2010). Both increases and
decreases in gliogenesis have been reported and may be time and insult dependent
(Ratnayake, Quinn et al. 2012; Jarlestedt, Naylor et al. 2013). Here PRRSV infection was found
to reduce the number of new cells differentiating into neurons by approximately 25% in both
males and females. New neurons in the dentate have been shown to be necessary for
hippocampal function including pattern separation and learning and memory, and disruptions
could lead to cognitive deficits (Deng, Aimone et al. 2010; Villeda, Luo et al. 2011). The number
of newly divided microglia cells was consistent across both sex and treatment at 15% of BrdU+
cells. There was a very high density of microglia in the subgranular zone and within the granule
cell layer, but the absolute numbers were not quantified. We also stained for astrocytes using
GFAP, but were unable to clearly identify BrdU+/GFAP+ cells. Astrocytes densely populated
the subgranular zone and only had projections into the granule cell layer, but no cell bodies.
The astrocytes were so dense that even with capturing z-stack image sets it was difficult to
positively identify a BrdU+/GFAP+ cell and not rule out that two cells were in close proximity.
Nonetheless, , the data show that in healthy control piglets 80% of BrdU+ cells develope into
neurons, and 15% into microglia. This leaves only 5% undetermined. In piglets infected with
PRRSV, however, only 55% of the BrdU+ cells develop into neurons, and 15% into microglia.
This leaves 30% of the BrdU+ cells unidentified. The undetermined cells could be astrocytes,
oligodendrocytes, or undifferentiated cells. Further, characterization is needed.
Research with adult rodents has shown that peripheral immune activation can reduce
the survival of new neurons in the dentate gyrus (Ekdahl, Claasen et al. 2003). The majority of
neurogenesis studies in adult rodents have used males only. As many developmental disorders
have a higher incidence in one sex or the other, it is important to study both males and females.
86
Here we find that there is a sexual dimorphism in the number of surviving BrdU+ cells in the
dentate gyrus with males having more surviving cells than females. This dimorphism is also
seen in the dentate gyrus and CA1 region of the neonatal rat (Zhang, Konkle et al. 2008;
Bowers, Waddell et al. 2010). PRRSV infection causes a significant reduction of surviving cells
in males, but does not affect females. This suggests that either females are not as susceptible
to inflammation or there is a protective mechanism to combat these signals. Alternatively, at
postnatal day 4, male rats have significantly more microglia in the dentate gyrus than females
which could lead to sex differences in the inflammatory response (Schwarz, Sholar et al. 2012).
Studies in rodents have used both acute inflammatory stimuli, such as LPS, or systemic
infection with Escherichia coli in male mice during the early postnatal period. (Bland, Beckley et
al. 2010; Jarlestedt, Naylor et al. 2013) These studies show reductions in new cell survival in
the dentate gyrus, consistent with what we found in the male piglets.
In addition to neurogenesis, changes in hippocampal neuron morphology were
assessed. Dentate granule cells were characterized as they have been shown to be sensitive
to inflammatory insult (Jurgens, Amancherla et al. 2012). The soma location within the granule
cell layer may be associated with “age” of the neuron as new cells are born in the subgranular
zone and migrate into the inner granule cell layer (Mongiat and Schinder 2011). Although we
didn’t specifically test for cell age, the microenvironment may be different for the inner granule
cell neurons and inflammation may affect them differently (Mongiat and Schinder 2011).
PRRSV infection caused a shifting in the shape of the inner granule cell dendritic tree,
extending the primary dendrite length before the first branching points. These results are
dissimilar to adult influenza studies which found that inflammation in adulthood caused a
retraction of dendrites (Jurgens, Amancherla et al. 2012). The reason for the extension of the
primary dendrite is not clear. The highest density of microglia was found in the subgranular
zone, so it may be that the inner granule layer cells are exposed to a higher pro-inflammatory
cytokine load. Additionally, studies have found that size of the dentate increases with prenatal
87
inflammation, so it is possible that the primary dendrite must travel through more cells before
starting to branch (Golan, Lev et al. 2005).
Additionally, results showed that there was a sexual dimorphism in the complexity of
outer dentate granule cell neurons, but the shape of the dendritic tree was similar among males
and females. This difference is similar to rodents (Juraska, Fitch et al. 1985). We also found
that the outer dentate granule cell neurons were more complex than the inner neurons, also
similar to rodents (Green and Juraska 1985). Although no differences were found in dendritic
spine density, the distribution of spine morphology is noteworthy. During early development,
stubby spines are dominant with some filipodia (Nimchinsky, Sabatini et al. 2002). As synaptic
connections are made and strengthened, the spine takes on a more mature mushroom
morphology (Bourne and Harris 2008). Our data illustrate that the most abundant morphology
was stubby and mushroom spines, a similar trend to the postnatal distribution in rodents (Harris,
Jensen et al. 1992).
In conclusion, inflammation during early-life has the potential to cause short- and long-
term disruptions in brain development. Our data indicate that neonatal respiratory infection
reduces cell survival, changes cell fate, and alters hippocampal cell morphology. The
conclusions of this study are limited as changes were only quantified at one end point. The
PRRSV piglets were almost symptomatically recovered at this time. Tracking changes during
peak sickness and at long-term endpoints would be useful. Additional work is needed to
characterize the unidentified BrdU+ cells. Regardless of these limitations, we show that early-
life respiratory infection can impact brain development. Further work with this model will allow
for testing of therapeutic strategies to modulate the neuroimmune response with aims of
preventing adverse developmental outcomes.
88
4.6 Figures
Figure 4.1. The representative section shows an overview of the hippocampus (top) with a magnified view of the suprapyramidal blade of the dentate gyrus (bottom). BrdU+ cells are indicated by DAB staining. Magnification = 2.5X (top), 20X (bottom); scale bar, 100 μm.
89
Figure 4.2. A-B, Representative example of a Golgi-stained dentate gyrus neuron (A) with corresponding tracing and Sholl analysis (B). Magnification = 40X; scale bar 100 μm.
90
2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32
0
2
4
6
8
10
ControlPRRSV
A
* * * * *
** * *
Day of Age
Wei
gh
t (k
g)
7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 3137
38
39
40
41
B
**
**
** * * *
** *
* * **
* **
Day of Age
Te
mp
era
ture
( C
)
7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 301.5
2.0
2.5
3.0
3.5
C
** * * * *
** *
* *
Day of Age
Fe
ed
ing
Sc
ore
Figure 4.3. A-C, Daily recorded sickness measures. Body weight was recorded from the start of the experiment (A) and temperature (B) and feeding score (C) were measured starting prior
to inoculation at 7 days of age. Data presented are means ± SEM (*p < 0.05 compared to controls).
91
Male Female5
10
15
20
25
30ControlPRRSV
a
bb
b
AB
rdU
+ c
ells
/ m
m3
(x10
3 )
Male Female5
10
15
20
25
30ControlPRRSV
a
b
ab
B
Brd
U+
cel
ls /
mm
3(x
103 )
Male Female0
3
6
9
12
15ControlPRRSV
a
ab
b
ab
C
Brd
U+
cel
ls /
mm
3(x
103 )
Figure 4.4. A-C, Density of newly divided (BrdU+) cells in the suprapyramidal blade (A), infrapyramidal blade (B), and hilus (C) of the denate gyrus. Groups are separated by sex and treatment. Letters indicate the groups that were significantly different (p < 0.05).
92
Male Female0
20
40
60
80
100
ControlPRRSV
a a
bb
%
Brd
U-N
euN
/Brd
U
Male Female0
5
10
15
20
25
ControlPRRSV
%
Brd
U-I
BA
1/B
rdU
C D
A B
Figure 4.5. A-D, Representative sections showing double labeling (A) of antibodies against BrdU (new cell; red) and NeuN (mature neuron; green). Representative section showing triple labeling (B) with antibodies against BrdU (new cell; blue), Iba-1 (macrophage/microglia; red), and GFAP (astrocyte; green). The granule cell layer is located in the upper half of each photomicrograph with the hilar region located in the bottom half. The proportion of newly divided cells that express NeuN (C) and IBA-1 (D) are plotted by sex and treatment. Means ± SEM are plotted with letters indicating differences between groups (p < 0.05). Magnification = 20X; scale bar = 50 μm.
93
Inner DG Outer DG0
200
400
600
800
1000
MaleFemale
*
AT
ota
l D
end
riti
c L
eng
th (m
)
Inner DG Outer DG0
10
20
30
40
50
MaleFemale
*
B
To
tal I
nte
rsec
tio
ns
20 100
200
300
0
2
4
6 MaleFemale
Outer DG
C
Distance From the Soma ( m)
# o
f In
ters
ecti
on
s
20 100
200
300
0
1
2
3
4
5ControlPRRSV
** *
*
D
Inner DG
Distance From the Soma (m)
# o
f In
ters
ecti
on
s
Figure 4.6. A-D, Females had reduced total dendritic length (A) and total intersections (B) in the outer dentate granule (DG) cell layer. Sholl analysis revealed that females had fewer intersections in the outer DG (C) and that PRRSV causes shifting of the first branching points away from the soma in the inner DG (D). Data are presented as means ± SEM (*p < 0.05 compared to controls).
94
Inner Outer0.0
0.2
0.4
0.6
0.8
1.0ControlPRRSV
Sp
ines
/ m
Figure 4.7. PRRSV infection did not alter spine density of dendritic granule cells located in the inner or outer DG. Data are represented as means ± SEM.
95
Thin Stubby Mushroom Filopodium Branched0.0
0.2
0.4
0.6
Sp
ines
/ m
Thin Stubby Mushroom Filopodium Branched0.0
0.2
0.4
0.6
ControlPRRSVS
pin
es/
mA
B
Figure 4.8. A-B. PRRSV infection did not alter the distribution of spine morphology in the inner DG (A) or outer DG (B). Data are represented as means ± SEM.
96
4.7 References
Bland, S. T., J. T. Beckley, et al. (2010). "Enduring consequences of early-life infection on glial and neural cell genesis within cognitive regions of the brain." Brain Behav Immun 24(3): 329-338.
Boksa, P. (2010). "Effects of prenatal infection on brain development and behavior: A review of
findings from animal models." Brain, Behavior, and Immunity 24(6): 881-897. Bourne, J. N. and K. M. Harris (2008). "Balancing structure and function at hippocampal
dendritic spines." Annu Rev Neurosci 31: 47-67. Bowers, J. M., J. Waddell, et al. (2010). "A developmental sex difference in hippocampal
neurogenesis is mediated by endogenous oestradiol." Biol Sex Differ 1(1): 8. Cacci, E., M. A. Ajmone-Cat, et al. (2008). "In vitro neuronal and glial differentiation from
embryonic or adult neural precursor cells are differently affected by chronic or acute activation of microglia." Glia 56(4): 412-425.
Conrad, M. S., R. N. Dilger, et al. (2012). "Brain Growth of the Domestic Pig (Sus scrofa) from 2
to 24 Weeks of Age: A Longitudinal MRI Study." Dev Neurosci 34(4): 291-298. Dantzer, R. and K. W. Kelley (2007). "Twenty years of research on cytokine-induced sickness
behavior." Brain, Behavior, and Immunity 21(2): 153-160. Deng, W., J. B. Aimone, et al. (2010). "New neurons and new memories: how does adult
hippocampal neurogenesis affect learning and memory?" Nat Rev Neurosci 11(5): 339-350.
Dietrich, R. B., W. G. Bradley, et al. (1988). "MR evaluation of early myelination patterns in
normal and developmentally delayed infants." American Journal of Roentgenology 150(4): 889-896.
Dilger, R. N. and R. W. Johnson (2010). "Behavioral assessment of cognitive function using a
translational neonatal piglet model." Brain Behav Immun 24(7): 1156-1165. Done, S. H., D. J. Paton, et al. (1996). "Porcine reproductive and respiratory syndrome (PRRS):
a review, with emphasis on pathological, virological and diagnostic aspects." Br Vet J 152(2): 153-174.
Duan, X., H. J. Nauwynck, et al. (1997). "Virus quantification and identification of cellular targets
in the lungs and lymphoid tissues of pigs at different time intervals after inoculation with porcine reproductive and respiratory syndrome virus (PRRSV)." Vet Microbiol 56(1-2): 9-19.
Ekdahl, C. T. (2012). "Microglial activation - tuning and pruning adult neurogenesis." Frontiers in
Pharmacology 3. Ekdahl, C. T., J.-H. Claasen, et al. (2003). "Inflammation is detrimental for neurogenesis in adult
brain." Proceedings of the National Academy of Sciences 100(23): 13632-13637.
97
Elmore, M. R., M. D. Burton, et al. (2014). "Respiratory viral infection in neonatal piglets causes
marked microglia activation in the hippocampus and deficits in spatial learning." J Neurosci 34(6): 2120-2129.
Elmore, M. R., R. N. Dilger, et al. (2012). "Place and direction learning in a spatial T-maze task
by neonatal piglets." Anim Cogn 15(4): 667-676. Golan, H. M., V. Lev, et al. (2005). "Specific neurodevelopmental damage in mice offspring
following maternal inflammation during pregnancy." Neuropharmacology 48(6): 903-917. Graciarena, M., A. M. Depino, et al. (2010). "Prenatal inflammation impairs adult neurogenesis
and memory related behavior through persistent hippocampal TGFβ1 downregulation." Brain, Behavior, and Immunity 24(8): 1301-1309.
Green, E. J. and J. M. Juraska (1985). "The dendritic morphology of hippocampal dentate
granule cells varies with their position in the granule cell layer: a quantitative Golgi study." Experimental Brain Research 59(3): 582-586.
Green, H. F. and Y. M. Nolan (2014). "Inflammation and the developing brain: consequences for
hippocampal neurogenesis and behavior." Neurosci Biobehav Rev. Green, H. F., E. Treacy, et al. (2012). "A role for interleukin-1beta in determining the lineage fate
of embryonic rat hippocampal neural precursor cells." Mol Cell Neurosci 49(3): 311-321. Hall, C. B., G. A. Weinberg, et al. (2009). "The Burden of Respiratory Syncytial Virus Infection in
Young Children." New England Journal of Medicine 360(6): 588-598. Harris, K. M., F. E. Jensen, et al. (1992). "Three-dimensional structure of dendritic spines and
synapses in rat hippocampus (CA1) at postnatal day 15 and adult ages: implications for the maturation of synaptic physiology and long-term potentiation." J Neurosci 12(7): 2685-2705.
Huttenlocher, P. R. (1979). "Synaptic density in human frontal cortex - developmental changes
and effects of aging." Brain Res 163(2): 195-205. Jarlestedt, K., A. S. Naylor, et al. (2013). "Decreased survival of newborn neurons in the dorsal
hippocampus after neonatal LPS exposure in mice." Neuroscience 253(0): 21-28. Juraska, J. M., J. M. Fitch, et al. (1985). "Sex differences in the dendritic branching of dentate
granule cells following differential experience." Brain Research 333(1): 73-80. Jurgens, H. A., K. Amancherla, et al. (2012). "Influenza Infection Induces Neuroinflammation,
Alters Hippocampal Neuron Morphology, and Impairs Cognition in Adult Mice." The Journal of Neuroscience 32(12): 3958-3968.
Kelley, K. W., R.-M. Bluthé, et al. (2003). "Cytokine-induced sickness behavior." Brain,
Behavior, and Immunity 17(1, Supplement 1): 112-118.
98
Kohman, R. A., E. K. DeYoung, et al. (2012). "Wheel running attenuates microglia proliferation and increases expression of a proneurogenic phenotype in the hippocampus of aged mice." Brain, Behavior, and Immunity(0).
Lynch, M. A. (2002). "Interleukin-1 beta exerts a myriad of effects in the brain and in particular in
the hippocampus: analysis of some of these actions." Vitam Horm 64: 185-219. Meyer, U., J. Feldon, et al. (2011). "Schizophrenia and autism: both shared and disorder-
specific pathogenesis via perinatal inflammation?" Pediatr Res 69(5 Pt 2): 26R-33R. Milatovic, D., S. Zaja-Milatovic, et al. (2003). "Pharmacologic suppression of neuronal oxidative
damage and dendritic degeneration following direct activation of glial innate immunity in mouse cerebrum." J Neurochem 87(6): 1518-1526.
Mongiat, L. A. and A. F. Schinder (2011). "Adult neurogenesis and the plasticity of the dentate
gyrus network." Eur J Neurosci 33(6): 1055-1061. Nimchinsky, E. A., B. L. Sabatini, et al. (2002). "STRUCTURE AND FUNCTION OF DENDRITIC
SPINES." Annual Review of Physiology 64(1): 313-353. Ratnayake, U., T. A. Quinn, et al. (2012). "Behaviour and hippocampus-specific changes in
spiny mouse neonates after treatment of the mother with the viral-mimetic Poly I:C at mid-pregnancy." Brain Behav Immun 26(8): 1288-1299.
Rice, D. and S. Barone, Jr. (2000). "Critical periods of vulnerability for the developing nervous
system: evidence from humans and animal models." Environ Health Perspect 108 Suppl 3: 511-533.
Richwine, A. F., A. O. Parkin, et al. (2008). "Architectural changes to CA1 pyramidal neurons in
adult and aged mice after peripheral immune stimulation." Psychoneuroendocrinology 33(10): 1369-1377.
Schwarz, J. M., P. W. Sholar, et al. (2012). "Sex differences in microglial colonization of the
developing rat brain." Journal of Neurochemistry 120(6): 948-963. Shin, J. H. and T. W. Molitor (2002). "Localization of porcine reproductive and respiratory
syndrome virus infection in boars by in situ riboprobe hybridization." J Vet Sci 3(2): 87-96.
Tashiro, A. and R. Yuste (2003). "Structure and molecular organization of dendritic spines."
Histol Histopathol 18(2): 617-634. Tong, L., R. Balazs, et al. (2008). "Interleukin-1 beta impairs brain derived neurotrophic factor-
induced signal transduction." Neurobiol Aging 29(9): 1380-1393. Vallieres, L., I. L. Campbell, et al. (2002). "Reduced hippocampal neurogenesis in adult
transgenic mice with chronic astrocytic production of interleukin-6." J Neurosci 22(2): 486-492.
Villeda, S. A., J. Luo, et al. (2011). "The ageing systemic milieu negatively regulates
neurogenesis and cognitive function." Nature 477(7362): 90-94.
99
Wu, M. D., A. M. Hein, et al. (2012). "Adult murine hippocampal neurogenesis is inhibited by
sustained IL-1beta and not rescued by voluntary running." Brain Behav Immun 26(2): 292-300.
Wu, M. D., S. L. Montgomery, et al. (2013). "Sustained IL-1β expression impairs adult
hippocampal neurogenesis independent of IL-1 signaling in nestin+ neural precursor cells." Brain, Behavior, and Immunity 32(0): 9-18.
Zhang, J. M., A. T. Konkle, et al. (2008). "Impact of sex and hormones on new cells in the
developing rat hippocampus: a novel source of sex dimorphism?" Eur J Neurosci 27(4): 791-800.
100
CHAPTER 5
MINOCYCLINE ADMINISTRATION DOES NOT ATTENUATE MICROGLIA ACTIVATION AND NEUROINFLAMMATION INDUCED BY PORCINE REPRODUCTIVE AND
RESPIRATORY SYNDROME VIRUS
5.1 Abstract
Microglia cells are the resident macrophages of the brain and are important for their
immunological response as well as maintaining normal tissue homeostasis. Through immune-
to-brain communication, microglia provide a “sixth sense,” monitoring the immunological state of
the periphery and allow for the propagation and amplification of inflammatory signals to the
central nervous system. Activation of microglia during the neonatal period has the potential to
disrupt many developmental processes. Porcine reproductive and respiratory syndrome virus
(PRRSV) is a respiratory viral infection which causes microglia activation and
neuroinflammation in piglets. Using a neonatal piglet model with PRRSV infection, we tested
the efficacy of using a chronically administered, high dose of minocycline to prevent
neuroinflammation. Minocycline is a second generation tricyclic antibiotic that has been shown
to prevent microglia activation and reduce neuroinflammation in vitro and in rodent models.
Minocycline did not rescue weight gain changes or prevent febrile response in the PRRSV
piglets and caused a reduction in feeding score in the PRRSV animals. Minocycline treatment
failed to reduce microglia activation in the hippocampus of PRRSV piglets and induced
activation in control piglets. Additionally, minocycline treatment caused a significant increase in
IL-1β and a trend towards increased TNF-α in the hippocampus. PRRSV piglets also had a
significant reduction in BDNF. Overall, these data indicate that high dose minocycline does not
attenuate neuroinflammation due to PRRSV. The high dose, commonly used in rodent models,
may itself lead to neuroinflammation in the neonatal period possibly due to increased
intracranial hypertension or bilirubin-induced brain damage.
101
5.2 Introduction
Microglia have been shown to have a multifaceted role in the central nervous system
including maintenance of normal tissue homeostasis, response to inflammation and pathogens,
and normal brain development (Harry 2013). Microglia are able to “sense” the inflammatory
state of the periphery through immune-to-brain communication and become activated in
response to peripheral inflammation (Dantzer, Konsman et al. 2000). This bi-directional
communication allows modification of physiology and behavior in response to injury or infection.
Upon activation, microglia can create a neuroinflammatory environment through production of
prostaglandins, free radicals, and pro-inflammatory cytokines (Perry and Gordon 1988; Gordon
2003). Much work has been done to characterize the impact of inflammation on brain function
and behavior in adults, but much less is known about how it can affect brain growth and
development. Immune activation during the neonatal time period may disrupt synaptogenesis,
dendritic growth, gliogenesis, and myelination, which are all sensitive to inflammation (Rice and
Barone 2000; Richwine, Parkin et al. 2008; Schmitz and Chew 2008; Bitzer-Quintero and
Gonzalez-Burgos 2012).
Porcine reproductive and respiratory syndrome virus (PRRSV) is an enveloped, positive-
sense, single-stranded RNA virus that belongs to the Arteriviridae family within the order
Nidovirales (Done, Paton et al. 1996). PRRSV causes interstitial pneumonia in piglets by
infecting the mononuclear myeloid cells in the lungs and activates microglia either through
immune-to-brain signaling pathways or by entering the brain (Done, Paton et al. 1996). In a
recent study, PRRSV infection caused elevated expression of pro-inflammatory genes, including
interferon-γ, TNF-α, and IL-1β, in several brain regions and caused elevated peripheral pro-
inflammatory cytokines 20 d after inoculation with PRRSV (Elmore, Burton et al. 2014).
Furthermore, 82% and 43% of microglia cells were activated (MHC II+) 13 and 20 d after
inoculation (Elmore, Burton et al. 2014). Microglial activation was positively correlated with
102
fever, and negatively correlated with learning and memory and food motivation (Elmore, Burton
et al. 2014).
We previously found PRRSV infection in the neonatal piglet to have profound effects on
brain development. Using an MRI-based approach, PRRSV infection caused reductions in grey
and white matter in multiple brain regions including the primary visual cortex (Chapter 2).
PRRSV infection may have caused white matter disruptions in the corpus callosum and also
reduced metabolite concentration in the hippocampus. In a separate but similar study, we found
PRRSV infection to inhibit neurogenesis by ~25% and alter dendritic architecture of granule cell
neurons. Collectively, these findings suggest a close relationship between infection, microglia
activation, neuroinflammation, and neurodevelopment. As alterations in neuroimmune activity
have been proposed to underlie psychiatric disorders including autism and schizophrenia, it
might be useful to inhibit microglial cells during infection at critical stages of development
(Meyer, Feldon et al. 2011).
Minocycline is a second-generation tetracycline antibiotic that has been shown to
have neuroprotective effects (Kim and Suh 2009). Minocycline is able to cross the blood-brain
barrier where it exerts both an anti-inflammatory and anti-apoptotic action. These effects are
mediated by reductions in microglia activation through a p38 MAPK-dependent mechanism
(Tikka and Koistinaho 2001). Minocycline has shown both protective and deleterious results as
a therapeutic in rodent models of hypoxia-ischemia, multiple sclerosis, Parkinson’s disease, and
Alzheimer’s disease (Yrjanheikki, Keinanen et al. 1998; Yang, Sugama et al. 2003; Casarejos,
Menendez et al. 2006; Choi, Kim et al. 2007; Chen, Pi et al. 2010). Multiple human clinical trials
have also been conducted with minocycline, with the results showing concern about the
translatability of rodent studies to humans (Blum, Chtarto et al. 2004; Kim and Suh 2009). To
date, there is a significant gap of evaluation of minocycline treatment in larger animal models
which may be more predictive of human outcomes.
103
Therefore, the goal of the current study was to evaluate chronic minocycline
administration as a potential therapeutic for neuroinflammatory changes due to PRRSV
infections. Changes in sickness symptoms, microglia activation, and hippocampal gene
expression are evaluated.
5.3 Materials and Methods
Animals, Housing, and Feeding
Naturally farrowed, crossbred piglets from three separate litters (12 females and 12
males) were obtained from the University of Illinois swine herd. Piglets were brought to the
biomedical animal facility on PD 2 (to allow for colostrum consumption from the sow) and placed
in individual cages (0.76 m L × 0.58 m W × 0.47 m H) designed for neonatal piglets (Elmore,
Burton et al. 2014). Each cage was positioned in a rack, with stainless steel perforated side
walls and clear acrylic front and rear doors within one of two separate but identical disease
containment chambers that have been described (Elmore, Burton et al. 2014). Each cage was
fitted with flooring designed for neonatal animals (Tenderfoot/NSR, Tandem Products, Inc.,
Minneapolis, MN, USA). A toy (plastic Jingle BallTM, Bio-Serv, Frenchtown, NJ, USA) was
provided to each piglet. Room temperature was maintained at 27ºC and each cage was
equipped with an electric heat pad (K&H Lectro-KennelTM Heat Pad, K&H Manufacturing, LLC,
Colorado Springs, CO, USA). Piglets were maintained on a 12-h light/dark cycle; however,
during the dark cycle minimal lighting was provided.
Piglets were fed a commercial sow milk replacer (Advance Liqui-Wean, Milk Specialties
Co., Dundee, IL, USA). Milk was reconstituted daily to a final concentration of 206 g/L using tap
water and supplied at a rate of 285 mL/kg BW (based on daily recorded weights) to a stainless
steel bowl via a peristaltic pump (Control Company, Friendswood, TX). Using this automated
104
feeding system (similar to that described previously (Dilger and Johnson 2010)), piglets
received their daily allotted milk over 18 meals (once per hour). All animal experiments were in
accordance with the National Institute of Health Guidelines for the Care and Use of Laboratory
Animals and approved by the University of Illinois at Urbana-Champaign Institutional Animal
Care and Use Committee.
Experimental Design and Treatments
Upon arrival, piglets were assigned to groups based on sex, litter of origin, and body
weight. A 2 × 2 factorial arrangement with the variables of ± PRRSV and ± minocycline was
used with three males and three females per group. At PD 7, 24 piglets (12 males and 12
females) were inoculated intranasal with either 1mL of 1x10^5 50% tissue culture infected dose
(TCID 50) of live PRRSV (strain P129-BV, obtained from the School of Veterinary Medicine at
Purdue University, West Lafayette, IN, USA) or sterile phosphate buffered saline (PBS).
Piglets in the minocycline group were administered one 100 mg capsule of minocycline
hydrochloride orally every day starting at PD 7 (Ranbaxy Pharmaceuticals, Jacksonville, FL,
USA). Minocycline was administered 2 hours prior to the first feeding using a cat piller. The cat
piller allows for non-invasive administration of capsules in larger animals. Controls were
subjected to similar stress of capsule administration using the cat piller. Six male piglets were
removed from the study due to bacterial infection or weight loss due to diarrhea. All remaining
piglets were sacrificed at PD 28 [CON/CON: n = 5 (2M, 3F); CON/MINO: n = 5 (2M, 3F);
PRRSV/CON: n = 3 (0M, 3F); PRRSV/MINO: n = 5 (2M, 3F)].
105
Assessment of Infection
Daily body weights (kg) were obtained to monitor piglets’ growth. In addition, daily rectal
temperatures were obtained PD 7 through PD 28. The willingness of the piglets to consume
their first daily meal was determined from PD 7 to PD 28 using a feeding score (1 = no attempt
to consume the milk; 2 = attempted to consume the milk, but did not finish within 1 min; 3 =
consumed all of the milk within 1 min).
The presence of PRRSV antibodies in the serum of all piglets at the end of the study
was analyzed by the Veterinary Diagnostic Laboratory (University of Illinois, Urbana, Illinois)
using a PRRSV-specific ELISA kit (IDEXX Laboratories). This assay has 98.8% sensitivity and
99.9% specificity, with an S/P ratio of >0.4 indicating a positive sample.
Euthanasia and Tissue Collection
All piglets were euthanized at PD 28 by first anesthetizing using a
telazol:ketamine:xylazine solution (50.0 mg of tiletamine, plus 50.0 mg of zolazepam,
reconstituted with 2.50 mL ketamine [100 g/L] and 2.50 mL xylazine [100 g/L]; Fort Dodge
Animal Health, Fort Dodge, IA, USA) administered at 4.4mg/kg BW. Piglets were euthanized
via intracardiac (i.c.) administration of sodium pentobarbital (86.0 mg/kg of BW; Fatal Plus®,
Vortech Pharmaceuticals, Dearborn, MI, USA). Brains were extracted, weighed, and dissected
for tissue collection. Brain tissues from the left hemisphere were collected, rinsed in PBS, and
stored at -80ºC in RNA later (Ambion, Carlsbad, CA, USA) until processing. The right
hippocampus was used for microglia isolation.
106
Microglia Isolation and Flow Cytometry
To assess the activational state, microglia were isolated and then characterized using
flow cytometry as previously reported (Elmore, Burton et al. 2014). Briefly, the right
hippocampus was collected from each piglet and rinsed in PBS. Tissues were homogenized in
DPBS (Sigma, St. Louis, MO, USA) with 0.2% glucose by pushing them through a 70 μm nylon
mesh strainer (BD Biosciences, San Jose, CA, USA). The cells were spun at 600 x g for 6 min
and the supernatant removed and resuspended with 70% isotonic Percoll (GE Healthcare,
Waukesha, WI, USA). A discontinuous gradient with 50%, 35%, and 0% isotonic Percoll was
overlaid and spun for 20 min at 2000 x g. The microglia were collected from the interface of the
50% and 70% layers and spun down to form a pellet. The pellet was resuspended in flow buffer
(PBS with 1% bovine serum albumin, Fisher, Waltham, MA, USA), 0.1% sodium azide (Sigma),
and 20 mM glucose (Sigma). Fc receptors were blocked with purified CD16/CD32 antibodies
(eBiosciences, CA, USA). The microglia were incubated with CD11b (Biolegend, San Diego,
Ca, USA), CD45 (AbD Serotec, Raleigh, NC, USA), and MHC II (Antibodies Online, Atlanta, GA,
USA). Using a LSR II Flow Cytometer (BD Biosciences), microglia were gated based on size,
granularity, and side scatter, as well as being CD11b+/CD45Int. Activated microglia were
characterized as being MCH II-positive.
Real Time PCR
RNA was extracted from approximately 50 mg of the left hippocampus of each piglet
using the Tri Reagent protocol (Sigma). cDNA synthesis was conducted with a QuanTect
Reverse Transcription Kit (Qiagen, Valencia, CA, USA) with removal of genomic DNA. Total
RNA quantity and purity was assessed with a NanoDrop system (Thermo Scientific, Wilmington,
DE, USA). Sequences for porcine BDNF, IL-10, IL-1β, IL-6, TGF-β, and TNF-α have been
107
previously reported and ribosomal protein L 19 (RPL19) was used as the housekeeping gene
(Elmore, Burton et al. 2014). Real-time PCR was conducted according to the Life Technologies
(Grand Island, NY, USA) Taqman Assay on Demand Gene Expression protocol. All PCRs were
plated in triplicate on a 384-well RT-PCR plate. An ABI PRISM 7900HT system (Life
Technologies) was used with standard cycle parameters. Data was analyzed using the
comparative threshold cycle (Ct) method (Livak and Schmittgen 2001) and are expressed as
relative changes in mRNA expression compared to controls.
Statistical Analysis
Data were analyzed using the MIXED procedure of SAS (SAS Institute, Cary, NC, USA).
For the sickness symptom measures, a three-way (PRRSV × Minocycline × Day) repeated
measures ANOVA with pig as the random effect was used. All other measures were analyzed
using a two-way (PRRSV × Minocycline) ANOVA. Means were compared using the least
significant difference (LSD) option in SAS and are presented as means ± SEM.
5.4 Results
Sickness Measures
Serum ELISA results indicated all control piglets were negative for PRRSV at the end of
the study, whereas all piglets inoculated were positive for PRRSV. Sickness responses for the
piglets with PRRSV and minocycline were tracked by measuring body weight gain, temperature,
and feeding scores (Figure 5.1). There was a significant main effect of day for body weight (F
(26, 364) = 20.29, p < 0.001) indicating that all pigs increased weight of the study period. There
was also a significant day × PRRSV interaction (F (26, 364) = 2.24, p < 0.001) showing that
108
PRRSV pigs had reduced weight gain compared to controls. There was no impact of
minocycline treatment on weight gain (F (1, 14) = 0). Rectal temperatures showed a significant
main effect of PRRSV (F (1, 14) = 11.46), p = 0.004) where PRRSV piglets had increased
temperatures (39.2 ± 0.2 °C) compared to controls (38.3 ± 0.2 °C). Feeding scores had a main
effect of day (F (20, 280) = 1.76, p = 0.02), minocycline (F (1, 14) = 9.05, p = 0.009), PRRSV (F
(1, 14) = 45.81, p < 0.001). In addition there were significant day × minocycline (F (20, 280) =
1.62, p = 0.05), day × PRRSV (F (20, 280) = 1.76, p = 0.02), minocycline × PRRSV (F (1, 14) =
9.05, p = 0.009), and a day × minocycline × PRRSV interaction (F (20, 280) = 1.62, p = 0.05).
The PRRSV animals had reduced feeding scores compared to controls, but this differed in
timing based on minocycline treatment. The PRRSV/CON piglets had reduced feeding scores
during the first week post-inoculation while the PRRSV/MINO piglets displayed a reduced
feeding score during the last two weeks of the study. Collectively, these data indicate infection
with PRRSV in the neonatal period induced clinical signs of illness.
Brain Weights
The whole excised brain weights were measured (Figure 5.2). There was a main effect
of PRRSV (F (1, 14) = 9.22, p = 0.009) where the PRRSV piglets had reduced brain weight
(42.1 ± 1.5 g) compared to controls (48.2 ± 1.3 g). There was also a trend for minocycline
treatment (F (1, 14) = 4.35, p = 0.06). When the raw brain weights were standardized to body
weight (brain weight/body weight), these differences disappeared (all p > 0.05) indicating that
these difference can be contributed to body size.
109
Microglia Activation
Cells were isolated from the hippocampus, identified as microglia (CD11b+ and CD45Int),
and activation was measured by MHC II using flow cytometry (Figure 5.3). The proportion of
activated microglia cells showed a significant effect of minocycline only (F (1, 14) = 11.86, p =
0.004) where minocycline piglets had increased activation (62%) compared to controls (19%).
Gene Expression
Gene expression changes in pro-inflammatory, anti-inflammatory, and neurotrophic
factors were measured (Figure 5.4). PRRSV piglets showed a decrease in BDNF expression
(0.57 ± 0.11) compared to controls (0.93 ± 0.10, F (1, 13) = 5.67, p = 0.03). For IL-1β, one
outlier was removed in PRRSV/MINO group. There was a significant effect of PRRSV (F (1, 12)
= 5.27, p = 0.04) where PRRSV piglets had increased expression (3.01 ± 0.46) compared to
controls (1.60 ± 0.41). There was also a significant effect of minocycline (F (1, 12) = 12.64, p =
0.004) where minocycline piglets (3.40 ± 0.41) had higher expression than controls (1.21 ±
0.46). There was a trend towards increased TNF-α in the minocycline groups (F (1,13) = 3.80,
p = 0.07). There were no significant differences found in IL-10, IL-6, and TGF-β expression (all
p > 0.05).
5.5 Discussion
Minocycline has potential for being a neuroprotective therapeutic for inflammatory
conditions in the central nervous system. Many studies have shown protective rolls of
minocycline in in vitro and in vivo acute inflammatory insults, but chronic administration
paradigms and testing in large animal models are lacking (Suk 2004; Yong, Wells et al. 2004).
110
Minocycline is thought to reduce neuroinflammation through a variety of mechanisms, but the
specific targets in the brain are being still being elucidated. Studies have shown that
minocycline has potent anti-inflammatory, anti-apoptotic, and antioxidant properties driven by
inhibition of p38 MAPK and inhibition of microglia activation (Plane, Shen et al. 2010). Here, we
tested whether chronic high dose minocycline could reduce microglia activation and
neuroinflammation caused by PRRSV in a neonatal piglet model. We found that minocycline
potentiated microglia activation and inflammatory cytokines in the hippocampus, contrary to
previous studies.
There is a substantial difference in dosage used in rodents and what has been deemed
safe for humans. Clinically, minocycline is routinely used to treat urinary tract infections and
acne with a dose of 2-4 mg/kg/day. Oral administration is most common as it provides 95-100%
bioavailability (Saivin and Houin 1988). The majority of rodent studies use intraperitoneal
injections and higher doses. In rodent models of hypoxia ischemia, 45-90 mg/kg was found to
be the neuroprotective dosage range, but this appears to be species specific where it is
protective in rats, but exacerbates damage in mice (Buller, Carty et al. 2009). For this study,
we chose a higher minocycline dose than clinically used in humans, similar to the rodent
neuroprotective range. One 100 mg tablet was administered daily starting at 7 days of age.
The effective dosing was on a sliding scale based on body weight grain where the average dose
at 7 days of age was 46 mg/kg/day and dropped to 17 mg/kg/day by the conclusion of the study.
Monitoring tolerance of this dose was important as many side effects have been
reported with high doses in humans. The most common adverse side effects include
gastrointestinal irritability, vestibular toxicity, and central nervous system disturbances (Buller,
Carty et al. 2009). Intracranial hypertension can be caused by minocycline and can lead to
multiple clinical manifestations (Lander 1989). We did not observe any gastrointestinal
disturbances in the control minocycline group. The PRRSV minocycline group did develop
111
diarrhea, but this is a common manifestation of PRRSV infection. The diarrhea was not
qualitatively different than the PRRSV control group. We did observe gait ataxia and hypertonic
limbs in two control minocycline and one PRRSV minocycline animal. The ataxia lasted for
roughly 3 days in each animal and occurred roughly 2-weeks after the start of minocycline
administration. The ataxia may be a sign of intracranial hypertension as it has been
documented before in humans (Round and Keane 1988).
Minocycline is not typically used during pregnancy and for infants as it is an ion chelator
that binds to calcium phosphate and is incorporated into developing teeth and bone (Buller,
Carty et al. 2009). We observed this in our minocycline piglets during brain extraction as the
skull had marked yellow coloration from minocycline (Figure 5.5). Additionally, tetracyclines can
disrupt binding of bilirubin to albumin leading to kernicterus, or bilirubin-induced brain damage
(Buller, Carty et al. 2009). Clinical features of kernicterus include reduced feeding, seizures,
and hypertonicity (Dennery, Seidman et al. 2001). Kernicterus may be a contributing factor in
this study as we observed hypertonic limbs in piglets (see above) and the PRRSV minocycline
group had reduced feeding scores beginning after one week of minocycline administration.
Kernicterus has also been shown to cause neuroinflammation and damage neurons,
astrocytes, oligodendrocytes, and microglia (Watchko and Tiribelli 2013). Unconjugated
bilirubin activates microglia through a p38 MAPK and ERK1/2 pathway which leads to increased
expression of TNF-α, IL-1β, and IL-6 (Silva, Vaz et al. 2010). This may explain the results found
in this study. Minocycline treatment significantly increased microglia activation in both the
control and PRRSV animals. Pro-inflammatory gene expression, including IL-1β and a trend in
TNF-α, were observed in the minocycline piglets. Although we did not observe jaundice in the
minocycline piglets, chronic exposure to even low levels of unconjugated bilirubin can lead to
neuroinflammation.
112
Use of minocycline in other large animal models of neuroinflammation have not shown
protective effects of minocycline, indicating differences between large animal and rodent
studies. Kay et al. used CLN6 South Hampshire sheep as a model for neuronal ceroid
lipofuscinoses (NCLs, or Batten disease) which is a fatal neurodegenerative disease (Kay and
Palmer 2013). They found that chronic high dose oral minocycline was successful in delivering
the drug to the CNS, but it did not inhibit microglia activation, astrocytosis, or neuronal loss.
Limitations to this study are the limited animal number, especially in the PRRSV/CON
group. Only 3 females were left in this group at the conclusion of the study impacting the ability
to include sex in statistical analysis. Nonetheless, the results of this study show the complexity
of minocycline use. Results obtained in rodent studies may not translate to large animal models
and may not be predictive of outcomes in human patients. More precise knowledge of the
mechanisms of minocycline in the brain of large animals would be beneficial for study design.
Finding the balance between having a high enough dose for neuroprotection, but low enough to
prevent side effects is key. High dose minocycline is not appropriate for neuroprotection in a
piglet PRRSV model, but additional research on other dosing regimens are warranted.
113
5.6 Figures
2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 280
2
4
6
8
Control Control
Control MinocyclinePRRSV Control
PRRSV Minocycline
* **
* * *
A
Day of Age
We
igh
t (k
g)
7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 2836
37
38
39
40
41
Control Control
Control MinocyclinePRRSV Control
PRRSV Minocycline
B
Day of Age
Tem
per
atu
re (C
)
7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 271.0
1.5
2.0
2.5
3.0
Control Control
Control MinocyclinePRRSV ControlPRRSV Minocycline
ac
ac ab
b
bc
bcbc
bc
bc bc
bc
bc
bc bc
C
Day of Age
Fir
st F
eed
ing
Sco
re
Figure 5.1. Sickness symptom measures. PRRSV piglets displayed reduced weight gain (A) and higher temperatures (B) than controls. The PRRSV piglets also showed decreases in the first feeding score (C). Data are represented as means ± SEM. (*<0.05 for PRRSV groups compared to CON; a = PRRSV/CON is significantly less than the control groups; b =
114
PRRSV/MINO is significantly less than the control groups; c = the PRRSV/CON and PRRSV/MINO are significantly different).
Control C
ontrol
Control M
inocy
clin
e
PRRSV Contro
l
PRRSV Min
ocycl
ine
35
40
45
50
55
60
A *
Bra
in W
eig
ht (
g)
Control C
ontrol
Control M
inocy
clin
e
PRRSV Contro
l
PRRSV Min
ocycl
ine
0
5
10
15
B
Bra
in W
eig
ht /
Bo
dy
We
ight
Figure 5.2. Brain weights. The PRRSV piglets had a significantly lower brain weight than controls (A). When standardized by brain weight, this difference disappeared (B). Data are represented as means ± SEM. (* < 0.05).
115
Control C
ontrol
Control M
inocy
clin
e
PRRSV Contro
l
PRRSV Min
ocycl
ine
0.0
0.2
0.4
0.6
0.8
1.0
MH
C I
I +
Ce
lls, %
Figure 5.3. Microglia activation. The minocycline treatment groups had significantly higher MHC-II expression than controls (p = 0.002). Data are represented as means ± SEM.
116
BDNFIL
-10
IL1- IL
-6
TGF-
TNF-
0
2
4
6 Control ControlControl MinocyclinePRRSV ControlPRRSV Minocycline
Rel
ativ
e E
xpre
ssio
n
Figure 5.4. Relative abundance of mRNA in the hippocampus. The PRRSV piglets had significantly lower expression of BDNF (p = 0.03). There were also significant main effects of PRRSV (p = 0.04) and minocycline (p = 0.004) for IL-1β. No significant differences were found in IL-10, IL-6, TGF-β, or TNF-α. Data are represented as means ± SEM.
117
Figure 5.5. Minocycline incorporation into the skull. Minocycline incorporation into the skull during ossification can be seen by comparing a control piglet skull (A) to a yellow tinted minocycline piglet skull (B). An additional view with the cuts through the skull show the extent of coloring caused by minocycline (C).
118
5.7 References
Bitzer-Quintero, O. K. and I. Gonzalez-Burgos (2012). "Immune system in the brain: a modulatory role on dendritic spine morphophysiology?" Neural Plast 2012: 348642.
Blum, D., A. Chtarto, et al. (2004). "Clinical potential of minocycline for neurodegenerative
disorders." Neurobiol Dis 17(3): 359-366. Buller, K. M., M. L. Carty, et al. (2009). "Minocycline: A neuroprotective agent for hypoxic-
ischemic brain injury in the neonate?" Journal of Neuroscience Research 87(3): 599-608.
Casarejos, M. J., J. Menendez, et al. (2006). "Susceptibility to rotenone is increased in neurons
from parkin null mice and is reduced by minocycline." J Neurochem 97(4): 934-946. Chen, X., R. Pi, et al. (2010). "Combination of methylprednisolone and minocycline
synergistically improves experimental autoimmune encephalomyelitis in C57 BL/6 mice." J Neuroimmunol 226(1-2): 104-109.
Choi, Y., H. S. Kim, et al. (2007). "Minocycline attenuates neuronal cell death and improves
cognitive impairment in Alzheimer's disease models." Neuropsychopharmacology 32(11): 2393-2404.
Dantzer, R., J. P. Konsman, et al. (2000). "Neural and humoral pathways of communication
from the immune system to the brain: parallel or convergent?" Auton Neurosci 85(1-3): 60-65.
Dennery, P. A., D. S. Seidman, et al. (2001). "Neonatal hyperbilirubinemia." N Engl J Med
344(8): 581-590. Dilger, R. N. and R. W. Johnson (2010). "Behavioral assessment of cognitive function using a
translational neonatal piglet model." Brain Behav Immun 24(7): 1156-1165. Done, S. H., D. J. Paton, et al. (1996). "Porcine reproductive and respiratory syndrome (PRRS):
a review, with emphasis on pathological, virological and diagnostic aspects." Br Vet J 152(2): 153-174.
Elmore, M. R., M. D. Burton, et al. (2014). "Respiratory viral infection in neonatal piglets causes
marked microglia activation in the hippocampus and deficits in spatial learning." J Neurosci 34(6): 2120-2129.
Gordon, S. (2003). "Alternative activation of macrophages." Nat Rev Immunol 3(1): 23-35. Harry, G. J. (2013). "Microglia during development and aging." Pharmacol Ther 139(3): 313-
326. Kay, G. W. and D. N. Palmer (2013). "Chronic oral administration of minocycline to sheep with
ovine CLN6 neuronal ceroid lipofuscinosis maintains pharmacological concentrations in the brain but does not suppress neuroinflammation or disease progression." J Neuroinflammation 10(1): 97.
119
Kim, H. S. and Y. H. Suh (2009). "Minocycline and neurodegenerative diseases." Behav Brain Res 196(2): 168-179.
Lander, C. M. (1989). "Minocycline-induced benign intracranial hypertension." Clin Exp Neurol
26: 161-167. Livak, K. J. and T. D. Schmittgen (2001). "Analysis of relative gene expression data using real-
time quantitative PCR and the 2(-Delta Delta C(T)) Method." Methods 25(4): 402-408. Meyer, U., J. Feldon, et al. (2011). "Schizophrenia and autism: both shared and disorder-
specific pathogenesis via perinatal inflammation?" Pediatr Res 69(5 Pt 2): 26R-33R. Perry, V. H. and S. Gordon (1988). "Macrophages and microglia in the nervous system." Trends
Neurosci 11(6): 273-277. Plane, J. M., Y. Shen, et al. (2010). "Prospects for minocycline neuroprotection." Archives of
Neurology 67(12): 1442-1448. Rice, D. and S. Barone, Jr. (2000). "Critical periods of vulnerability for the developing nervous
system: evidence from humans and animal models." Environ Health Perspect 108 Suppl 3: 511-533.
Richwine, A. F., A. O. Parkin, et al. (2008). "Architectural changes to CA1 pyramidal neurons in
adult and aged mice after peripheral immune stimulation." Psychoneuroendocrinology 33(10): 1369-1377.
Round, R. and J. R. Keane (1988). "The minor symptoms of increased intracranial pressure:
101 patients with benign intracranial hypertension." Neurology 38(9): 1461-1464. Saivin, S. and G. Houin (1988). "Clinical pharmacokinetics of doxycycline and minocycline." Clin
Pharmacokinet 15(6): 355-366. Schmitz, T. and L. J. Chew (2008). "Cytokines and myelination in the central nervous system."
ScientificWorldJournal 8: 1119-1147. Silva, S. L., A. R. Vaz, et al. (2010). "Features of bilirubin-induced reactive microglia: from
phagocytosis to inflammation." Neurobiol Dis 40(3): 663-675. Suk, K. (2004). "Minocycline suppresses hypoxic activation of rodent microglia in culture."
Neuroscience Letters 366(2): 167-171. Tikka, T. M. and J. E. Koistinaho (2001). "Minocycline provides neuroprotection against N-
methyl-D-aspartate neurotoxicity by inhibiting microglia." J Immunol 166(12): 7527-7533. Watchko, J. F. and C. Tiribelli (2013). "Bilirubin-Induced Neurologic Damage — Mechanisms
and Management Approaches." New England Journal of Medicine 369(21): 2021-2030. Yang, L., S. Sugama, et al. (2003). "Minocycline enhances MPTP toxicity to dopaminergic
neurons." J Neurosci Res 74(2): 278-285. Yong, V. W., J. Wells, et al. (2004). "The promise of minocycline in neurology." Lancet Neurol
3(12): 744-751.
120
Yrjanheikki, J., R. Keinanen, et al. (1998). "Tetracyclines inhibit microglial activation and are neuroprotective in global brain ischemia." Proc Natl Acad Sci U S A 95(26): 15769-15774.
121
CHAPTER 6
SUMMARY AND SIGNIFICANCE The aim of this research was to determine if neonatal respiratory viral infection induces
changes in the developing brain. Previously, we have shown that porcine reproductive and
respiratory viral infection (PRRS) causes marked microglia activation, increased pro-
inflammatory cytokines, and deficits in hippocampal dependent learning and memory. The
present work further expands upon this research to look at possible mechanisms for the deficits
in learning and memory and possible therapeutics.
Initial studies were conducted to develop magnetic resonance imaging techniques for
the neonatal piglet. Three works are presented in the appendices that allowed for studies in
chapter 3 to be conducted. First, basic methods were developed to measure regional brain
volumes in the piglet using structural MRI techniques, and these data are presented in appendix
A. Using this technique, a longitudinal study was conducted that characterized the regional
brain growth curves in the domestic pig (Appendix B). This study was significant as it allowed
for the comparison of the regional brain growth to humans and this confirmed the usefulness of
the piglet as an animal model for human brain development. Finally, an averaged brain and
MRI-based atlas was created for the neonatal piglet (Appendix C). This allowed for further
refinement of the structural MRI analysis including the ability to conduct voxel-based
morphometry studies. Additional method optimization and protocols were developed for the
piglet including diffusion tensor imaging and magnetic resonance spectroscopy.
Using these new methods, the results from Chapter 3 show that neonatal PRRS
infection does impact the macro development of the CNS. There were significant reductions in
gray and white matter, most notably in the primary visual cortex. This was the first time that
voxel-based morphometry has been used in a piglet model. There was also a trend for
reductions in fractional anisotropy in the corpus callosum, which suggest either delay or
disruption of myelination. Additionally, there were differences found in metabolite
122
concentrations within the hippocampus. Importantly, this research shows that MRI is an
important research tool for tracking brain growth and development, and is sensitive to insults
including infection. Further research is needed in the brain areas which had reductions in gray
and white matter volume in the PRRS piglets. Histological confirmation of this volume decrease
and characterization of the cellular changes are needed.
In Chapter 4, cellular changes due to PRRS infection were characterized in the
hippocampus. Focus was placed on two developmental processes within the dentate gyrus
which have been implicated in learning and memory. First, the dentate gyrus is one of two brain
areas which has postnatal neurogenesis. Bromodeoxyuridine administration, concurrent with
PRRS infection, allowed for the characterization of the impact of PRRS on new cell survival and
cell fate. A sexual dimorphism in new cell survival was found in which males had more
surviving cells than females. Additionally, only males had reduced survival of new cells with
PRRS infection. This shows that there may be vulnerability differences between the sexes, and
using male and female animals in research is important. Cell fate was also disrupted with
PRRS infection causing a decrease in the number of newly divided cells that form neurons. The
number of newly formed microglia were not affected by PRRS. Secondly, we found differences
due to sex and PRRS in dentate granule cell morphology. There were many similarities found
between rodents and pigs including males having more complex dendritic trees than females
and complexity differences due to location in the granule cell layer. PRRS infection caused a
morphological change in the inner granule cell neurons, with the initial sprouting point further
away from the cell soma. There were no changes in spine density or the distribution of the
morphology of the spines. These data show important changes at the cellular level caused by
early-life respiratory infection which may have implications for learning and memory.
Lastly, the studies in Chapter 5 were meant to study the mechanism of the changes in
the previous two chapters and evaluate minocycline as a possible therapeutic for
neuroinflammation. Minocycline is a second generation tetracycline antibiotic which prevents
123
the activation of microglia. Using minocycline for the duration of PRRS infection, this study
sought to prevent microglia activation in the PRRS piglets. The results indicate that minocycline
administration increased microglia activation, contrary to the majority of studies published in
other species. Minocycline may have caused a worsening of feeding scores in the PRRS
animals and also caused increased levels of pro-inflammatory gene expression in the
hippocampus. The results from this study may be due to a high dose, albeit neuroprotective in
rodents, that was used. High doses of minocycline have the potential to increase intracranial
pressure and cause bilirubin-induced brain damage. Thus, this shows that high dose
minocycline is not appropriate in the neonatal time period. Further research into finding a dose
that is high enough to exhibit neuroprotective effects, yet low enough to prevent side effects is
warranted. Collectively, this dissertation shows the early-life respiratory viral infection can effect
brain development. Finding better therapeutics to modulate the neuroimmune response could
prevent these changes and lead to better short- and long-term developmental outcomes.
124
APPENDIX A
MAGNETIC RESONANCE IMAGING OF THE NEONATAL PIGLET BRAIN
A.1 Abstract
Appeal for the domestic pig as a preclinical model for neurodevelopmental research is
increasing. One limitation, however, is lack of MRI methods for brain volume quantification in
the neonatal piglet. To address this void, anatomic MRI data (non-longitudinal) were acquired
from 2-week (n=6) and 5-week (n=6) old pigs using a three-dimensional T1-weighted
magnetization prepared gradient echo (MPRAGE) sequence on a MAGNETOM Trio 3T imager.
Manual segmentation was performed for volume estimations of total brain, cortical,
diencephalon, brainstem, cerebellar, and hippocampal regions. Hippocampal volumes were
also estimated by postmortem histological analysis using the Cavalieri method with planimetric
analysis. Strong correlations were found between the hippocampal volume estimates using
MRI and the histological section estimates of hippocampal volume, indicating accurate
volumetric measurements can be obtained using this MRI protocol. In addition to developing
and validating MRI methods for estimating brain volume, the present study also provides
evidence of substantial brain growth during this early period. The results indicate that MRI can
provide accurate estimates of brain region volume changes over time during the neonatal period
in piglets. The availability of a piglet model for use in longitudinal studies may be useful to
understanding effects of experimental factors on brain growth and development.
This material has been previously published. The copyright owner has provided permission to reprint. Conrad Matthew S, Dilger Ryan N, Nickolls Alec, Johnson Rodney W (2012) Magnetic resonance imaging of the neonatal piglet brain. Pediatr Res 71:179-84
125
A.2 Introduction
The domestic pig (Sus scrofa), due to its anatomic and physiologic similarities to
humans, is a common well accepted preclinical model in cardiovascular, metabolic, and
pediatric nutrition research (Miller and Ullrey 1987; Bellinger, Merricks et al. 2006; Dixon and
Spinale 2009). For the reason that pigs and humans also share similar patterns in brain growth
and development, appeal for the pig as a model for neurodevelopmental research has recently
increased (Pond, Boleman et al. 2000; Lind, Moustgaard et al. 2007). In both the pig and
human, the major brain growth spurt extends from late prenatal to early postnatal, which is
different from other common animal models (Dobbing and Sands 1979). Both pigs and humans
are born with a brain weight that is roughly 25% of the final adult weight (Dobbing and Sands
1979). Additionally, gross anatomical features including gyral pattern and distribution of grey
and white matter of the neonatal pig brain are similar to that of human infants (Dickerson and
Dobbing 1967; Thibault and Margulies 1998). Furthermore, due to the fact that the pig is a
precocial species with fairly well developed sensory and motor systems at birth, it is amenable
to behavioral testing to assess cognitive development as early as one to two weeks of age
(Dilger and Johnson 2010). Thus, the neonatal piglet is a potential preclinical translational
model for investigating the effects of early-life experiential factors (e.g., infection, nutrition) on
brain and cognitive development. As the piglet is already used to study parenteral and enteral
nutrition in preterm infants (Ganessunker, Gaskins et al. 1999), and preterm birth is associated
with a high rate of developmental delay and cognitive disability (Wood, Marlow et al. 2000), the
availability of the piglet for investigating brain and cognitive development would be particularly
welcome. One limitation, however, has been the ability to make use of quantitative magnetic
resonance imaging (MRI) for in vivo assessment of brain growth and development in neonatal
piglets.
Quantitative MRI is providing important information on brain development in early
childhood and adolescence (Pfefferbaum, Mathalon et al. 1994; Giedd, Vaituzis et al. 1996;
126
Giedd, Blumenthal et al. 1999; Gilmore, Lin et al. 2007; Knickmeyer, Gouttard et al. 2008) but
because of potential health concerns for infants, few studies have focused on the period from
birth to 4 years of age when dramatic brain development occurs. Most MRI studies of infants
have been with those born prematurely or with significant health complications (Beauchamp,
Thompson et al. 2008; Dubois, Benders et al. 2008; Thompson, Wood et al. 2008). Only
recently was MRI used to assess structural brain development of healthy full term infants from
birth to 2 years of age (Knickmeyer, Gouttard et al. 2008). The first year after birth, total brain
volume more than doubled and hemispheric grey and white matter increased 149% and 11%,
respectively. Although this application of MRI in human infants is impressive, the ability to
address mechanistic questions related to experiential influences on brain growth and
development cannot be readily done due to practical or ethical concerns.
In older pigs, it is possible to obtain reliable estimates of brain volumes using MRI
(Jelsing, Rostrup et al. 2005), and MRI has been used in experimental neurosurgical
procedures and to investigate several neuropathological conditions (Munkeby, De Lange et al.
2008; Rosendal, Pedersen et al. 2009; Björkman, Miller et al. 2010; Kuluz, Samdani et al. 2010;
Sandberg, Crandall et al. 2010). To our best knowledge, however, MRI has not been used to
obtain brain volumetric data for in vivo quantitative analysis in neonatal piglets. As a first step
toward addressing this void, here we report MRI scanning and analysis protocols for estimating
total brain, cortex, diencephalon, cerebellum, hippocampus, and brainstem volumes in neonatal
piglets at 2- and 5-weeks of age. The reliability of the MRI-based volume estimates was
assessed for the hippocampus by the Cavalieri method and planimetric analysis with physical
histological sections obtained postmortem. The results indicate that MRI can provide an
accurate in vivo estimate of brain region volume changes over time during the neonatal period.
Furthermore, substantial increases in volumes for all brain regions examined were evident
during the three-week time period. The potential to quantify brain growth and development in
127
vivo using MRI in a longitudinal study design makes the piglet an attractive preclinical animal
model.
A.3 Material and Methods
Subjects
A total of twelve pigs, six male and six female (Sus scrofa domestica, York breed), from
three litters, were obtained from the University of Illinois swine herd at two days of age and
placed into an artificial rearing system that has been previously described (Dilger and Johnson
2010). Briefly, each piglet was housed individually in an acrylic-sided cage with a 12 hour
light/dark cycle. Ambient temperature was maintained at 27˚C with intracage temperatures
maintained at 32˚C using radiant heaters located directly over the piglets. Piglets received a
commercially available milk replacer (Milk Specialties Global Animal Nutrition, Carpentersville,
IL) throughout the 5-week study. Milk replacer powder was reconstituted fresh each day and 50
mL of liquid diet was delivered to individual feeding bowls once an hour from 8:00 a.m. to
midnight using a custom, automated system (Dilger and Johnson 2010). The pigs had an
average weight of 1.86 kg ± 0.05 at birth and 3.08 kg ± 0.05 and 9.13 kg ± 0.34 at 2- and 5-
weeks of age, respectively. All procedures were in accordance with the National Institute of
Health Guidelines for the Care and Use of Laboratory Animals and approved by the University
of Illinois Institutional Animal Care and Use Committee.
Experimental Procedures
At 2-weeks of age, six of the pigs, three male and three female, were subjected to MRI
procedures. Immediately before scanning, pigs were anesthetized by intramuscular (i.m.)
injection of a telazol:ketamine:xylazine solution (TKX; 4.4 mg/kg body weight), and placed in a
prone position in the MRI scanner. The pigs remained anesthetized during the MRI scanning
procedure (approximately 20 minutes) and afterwards were euthanized by intracardiac
128
administration of an overdose of sodium pentobarbital (390 mg/ml Fatal Plus - administered at 1
ml/5 kg body weight). Brains were removed and a 2.5 cm3 tissue block encompassing the left
hippocampus was excised and placed in 4% paraformaldyhyde. Tissue blocks were stored in
fixative at 4°C for 3 weeks before processing. Tissue blocks from male and female littermates
were selected and processed for histological volume determination. The remaining six pigs
were scanned at 5-weeks of age after which they were euthanized and their brain tissue was
collected and processed as described above.
Imaging
Magnetic resonance scanning was conducted using a Siemens MAGNETOM Trio 3-T
imager by employing a Siemens 32-channel head coil (Erlangen, Germany). Anatomic images
were acquired using a three-dimensional T1-weighted magnetization prepared gradient echo
(MPRAGE) sequence with the following parameters: repetition time, TR = 1900 ms; echo time,
TE = 2.48ms; inversion time, TI = 900 ms, flip angle = 9˚, matrix = 256 x 256 (interpolated to
512 x 512), slice thickness = 1.0 mm. The final voxel size was 0.35 mm x 0.35 mm x 1.0 mm
across the entire head from the tip of the snout to the cervical/thoracic spinal cord junction.
Images were exported in the Digital Imaging and Communications in Medicine (DICOM) format
and analyzed using three-dimensional visualization software (AMIRA®, Visage Imaging, Inc.,
San Diego, CA).
In vivo Quantification of Brain Region Volume Using MRI
The volumes for total brain, cortex, diencephalon, cerebellum, hippocampus, and
brainstem were quantified using AMIRA® 3-D visualization software. Unsharp masking post-
processing was performed to improve grayscale inhomogeneities (Brinkmann, Manduca et al.
1998). Manual segmentation of the brain regions was facilitated by a Wacom Cintiq 21UX
graphic input screen and pen (Wacom, Vancouver, WA). The anatomical criterion for the
129
regions were based upon the stereotaxic atlas of Félix et al (1999). The boundaries for
segmentation are shown in Figure 1. For the hippocampus, the medial border was defined by
the choroidal fissure and the lateral border was the inferior horn of the lateral ventricle. The
superior border was the body of the lateral ventricle and the inferior border was
parahippocampal gyrus. The total brain volume was defined as the sum of all cortical,
subcortical, cerebellar, brainstem, and partial spinal cord volumes. The spinal cord was
segmented to the posterior boundary of the cerebellum. The volumes were calculated using the
material statistics function of AMIRA® which estimates the three-dimensional volume of the
manually segmented regions given the aforementioned voxel size. To determine the reliability
of these measures, segmentation was conducted by two independent observers and the intra-
and interobserver agreements calculated.
Quantification of Hippocampal Volume from Histological Sections
Paraformaldehyde-fixed hippocampal tissue was dehydrated and paraffin-embedded
using a Leica ASP300 tissue processor (Leica, Wetzlar, Germany). The tissue block was
sectioned in the axial plane at 5 µm intervals and every 10th section was collected. The serial
sections were hematoxylin and eosin (H&E) stained using a Sakura Tissue-Tek DRS H & E
stainer system and coverslipped with a Sakura Tissue-Tek Glas coverslipper (Sakura, Tokyo,
Japan). The slides were digitized at 20x magnification using a Nanozoomer Digital Pathology
System (Hamamatsu, Hamamatsu City, Japan), creating high resolution virtual slide images.
Digital images of the hippocampal region were exported from the virtual slides as JPEGs for
subsequent segmentation. Digital images of the serial sections were imported into AMIRA®
and the z-plane depth was adjusted to account for the 50 µm distance between collected slices.
Images were aligned and the hippocampus was manually segmented for each section. The
hippocampal region was defined as the dentate gyrus, CA1, and CA3 as shown in a
130
representative slide in Figure 2. After all tissue sections were segmented, the material statistics
function of AMIRA® was used to estimate hippocampal volume.
Statistics
For analysis of brain region volumes in 2- to 5-week old piglets, data were subjected to two-way
ANOVA (Age × Sex) using SAS software (SAS, Cary, NC). It was determined that there was
unequal variance between the data obtained for 2- and 5-week old piglets, and to correct for
this, data were log transformed before ANOVA. There was no significant effect of sex, therefore
the male and female data were combined for calculation of the means. Significance level was
set at p<0.05. Validation of MRI volume estimation of the hippocampus was based on the
technique employed in human studies by Bobinski et al. (Bobinski, de Leon et al. 2000). The
correlative relationship between hippocampal volumes obtained by MRI and histological
measurement was determined using GraphPad Prism 5 software (GraphPad Software, La Jolla,
CA). The Pearson product-moment correlation coefficient was used to determine the strength of
linear dependence between the two volume estimations.
The intraobserver and interobserver agreements for manual segmentation were
calculated based on statistical methods by Bland and Altman (Bland and Altman 1986). The
intraobserver agreement was calculated by the following formula where stx1 and ndx2 were
repeat measures of regional volumes of 6 pigs using manual segmentation by one rater
(M.S.C). This rater was blinded to the pig brain images being segmented and the order pig
brain images were segmented was randomized.
Intraobserver agreement index =
100
2/100
21
21
ndst
ndst
xx
xx
131
The interobserver agreement was calculated by the following formula where ax and bx are
measures of regional volumes of each pig using manual segmentation by two different raters
(M.S.C and A.N.).
Interobserver agreement index =
100
2/100
ba
ba
xx
xx
The intraobserver coefficient of variation (COV) was defined as 2 SD of the following equation
by the British Standards Institution (Institution 1998).
2/21
21
ndst
ndst
xx
xx
The interobserver COV was defined as 2 SD of the following equation
2/ba
ba
xx
xx
A.4 Results
Total brain and subregion volumes for 2- and 5-week old piglets (n = 6) are shown in
Table 1. As expected, there was a main effect of age on total brain, cortex, diencephalon,
cerebellum, hippocampus, and brainstem volumes, demonstrating considerable volume
increases from the 2- to 5-week period. There was no main effect of sex for any brain region
studied, indicating that the volumes were similar in males and females at this age (p > 0.05 for
all measures). Because of this, data for males and females (n = 6) were combined. Volume of
the left and right hippocampal formation was similar within each animal irrespective of sex or
age. Since no asymmetries were observed, the left and right hippocampal volumes were
combined to create a total hippocampal volume measure at each age. A three-dimensional
segmented brain including all of the aforementioned brain regions can be seen in Figure 3.
132
To determine the reliability of the MRI-based volume estimates, the volume of the
hippocampus was also determined from physical sections obtained postmortem following MRI.
The hippocampus was chosen because it was smallest of all areas examined and most difficult
to manually segment. Furthermore, technical challenges prevented whole brain slicing at 5µm
intervals. Using the histological sections, segmentation of individual hippocampal subregions
(dentate, CA1, and CA3) was possible, however, due to the small size of the hippocampus in
the young pig, it was not possible to identify these hippocampal subregions by MRI. Thus, only
whole hippocampal volume was compared between the two methods. To reduce variance due
to genetic variability, piglets from the same litter were used for the histological validation. A
total of 4 piglets were used, two that were 2-weeks of age and two that were 5-weeks of age.
The average histological-determined volumes at 2-weeks and 5-weeks of age were 165.29 mm3
(± 9.75) and 246.70 mm3 (± 11.74), respectively. The average MRI-determined volumes at 2-
and 5-weeks of age were 487.83 mm3 (± 38.28) and 658.87 mm3 (± 47.25), respectively. While
tissue blocks were not measured prior to processing, the difference in the MRI and histological-
determined volumes represents a reduction in the total volume (i.e. shrinkage) of approximately
64%. Finally, a strong correlation between hippocampal volumes determined by MRI and
histology was evident (r = 0.9878, P = 0.0122).
The median intraobserver and interobserver agreement for two independent raters for
MRI-based volume estimates is shown in table 2. Intraobserver agreement was consistent
across brain regions with the median values ranging from 94.4 to 96.6%. The coefficients of
variation ranged 2.1 to 6.4% across brain regions. The interobserver agreement medians
ranged from 96.6 to 98.8% with coefficients of variations ranging from 2.0 to 5.2%.
A.5 Discussion
MRI has been used in older pigs to quantify whole brain and sub-structure volumes, but
to our best knowledge, no studies have used MRI to characterize normal brain development
133
during the neonatal time period (Jelsing, Rostrup et al. 2005). Therefore, the current study
sought to determine if MRI could be used to estimate volumes of different brain regions in the
neonatal period. The important results showed that MRI and manual segmentation reliably
estimated volume changes over time of different brain regions. Furthermore, substantial
increases in volumes for all brain regions examined were evident during the three-week time
period.
Several challenges are presented when using MRI during the neonatal time period
including poor spatial resolution and low tissue contrast which causes difficulties discerning
between grey matter, white matter, and cerebrospinal fluid (Shi, Fan et al. 2010). Because of
this, using automated segmentation procedures in the neonate has been difficult. An additional
obstacle of structural analysis in the neonatal pig brain is the lack of a brain atlas specific to this
age period. Currently, the only MRI-based atlas for the pig is in adult animals, which is not
suitable for neonatal brain analysis (Saikali, Meurice et al. 2010). Because there is currently no
atlas available for pigs during the neonatal time period, we chose to use manual segmentation
for volume analysis. This technique requires significantly more time than semi- and fully
automated segmentation methods, but provides accurate volume estimations. Additional
imaging data sets are needed for creation of structural MRI atlases, which will allow for more
complex, automated segmentation methods. A future goal is to utilize more sophisticated
techniques used for human studies, such as neonatal cortical mapping, once aged-based MRI-
atlases have been created in the piglet (Dubois, Benders et al. 2008). Nonetheless, the current
procedure can be used in a longitudinal experimental design to assess the effects of
experiential factors (e.g., nutrition, infection) on brain growth and development. This is important
because similar to humans (Dobbing and Sands 1979), overall brain weight of piglets in the
neonatal period is about 25-35% of normal adults.
Studies employing MRI for volumetric analysis of specific brain regions typically
validated this modality using classic histological techniques and the Cavalieri method (Jelsing,
134
Rostrup et al. 2005). Here, we used a digital planimetry method with high-resolution segment
image sets from both MRI and histological sections. The planimetry method is a reliable
method for volume estimation in which areas of interest are outlined on a computer screen and
the included pixels in series of images is used for volume calculation (Cotter, Miszkiel et al.
1999). Using this method, we demonstrated a high significant correlation between hippocampal
volume determined from the histological sections and hippocampal volume determined using
MRI, showing MRI can provide accurate relative differences over time.
In addition to developing MRI methods for estimating brain volume, by examining 2-
week and 5-week old piglets, the present study also provides evidence of substantial brain
growth during this early three-week period. A limitation of the present study was that it was not
a longitudinal design due to the need to obtain brain sections for estimation of hippocampal
volume using the Cavalieri method. Nonetheless, the volumes for cortex and subcortical
regions including the hippocampus, diencephalon, cerebellum, and brainstem were 30-43%
greater in 5-week old piglets compared to 2-week old piglets. No main effect of sex on absolute
volumes or differences in volumes between 2- and 5-week olds was evident for these brain
regions. These data are similar to a study of humans aged 3 months to 30 years which found
no effects of sex on grey and white matter volumes after accounting for head size (Pfefferbaum,
Mathalon et al. 1994).
Additional limitations of this study include the simplicity of the volumetric analysis
technique and the histological methods for validation of the MRI volumes. Due to lack of age
specific MRI-based atlases in the neonatal piglet, more sophisticated regional volume
estimation methods could not be used. Our lab is collecting MRI data to construct these
averaged brain MRI-atlases in order to utilize techniques such as voxel-based morphology and
deformation-based morphology which have been shown to be powerful methods for volumetric
analysis in the neonatal brain (Boardman, Counsell et al. 2006). In addition, the automated
scull stripping features of many semi- and fully-automated procedures that were developed for
135
humans do not currently work with the pig due to anatomical differences. For this study, only
two independent observers were evaluated to construct the coefficients of variation (COV).
Although it would be ideal to evaluate additional observers, we obtained low COVs for our
methods, and the use of only two observers is similar to other manual segmentation protocols
for human brain region analysis (Bokde, Teipel et al. 2005; Yu, Zhang et al. 2010). For the
histological validation, there were differences in the absolute volumes between the MRI and
histologically-determined hippocampal volumes. These differences were due to the dehydration
and paraffin embedding process of the hippocampal tissue. We chose this approach in order to
be able to slice the tissue at the small intervals used in order to increase the resolution of our
histological method. Because of this, the correlation analysis is based on relative differences in
volume rather than absolute volumes. In addition, another limitation is that tissues from only
four brains were used for the correlation analysis, although all of these brains were from
littermate pigs allowing for reduction in variability. Despite the limitations of this study, the data
supports this as a reliable method for quantification of brain region volumes in the neonatal
piglet and future work will allow for more semi- and fully-automated segmentation of the piglet
brain.
In summary, reliable estimates of brain volume changes over time in neonatal piglets
can be obtained using MRI and manual segmentation procedures. Because the pig is a
gyrencephalic species thought to have brain growth and development similar to humans, it may
serve as a preclinical translational model for studying the developmental origins of
neuropsychiatric diseases. Using the methods introduced in this paper, ongoing studies will
characterize normal brain development in the pig from the neonatal period to adulthood in a
longitudinal design.
136
A.6 Figures and Tables
Figure A.1. MRI segmentation boundaries. The left (yellow) and right (red) hippocampus are segmented in the coronal, axial, and sagittal planes of view. The cortex (green), diencephalon (teal), brainstem (dark blue), and cerebellum (pink) are also shown.
137
Figure A.2. Hippocampus segmentation criteria for histological tissue sections. The outlined area includes the dentate gyrus (DG), CA1, and CA3 regions of a horizontal cross-sectional slice of hippocampal tissue from the dorsobasal portion of the hippocampus. The subiculum (Sub) was not included in volume analysis. Image was taken at 12.5X magnification.
138
Figure A.3. Three-dimensional brain reconstruction. The 3-D reconstruction shows the anatomical position of the left hippocampus (yellow) and right hippocampus (red) in relation to the cortex (green), diencephalon (teal), brainstem (dark blue), and cerebellum (pink).
139
Region 2 Week
Old 5 Week
Old %
Increase Age
mm3 (SE) mm3 (SE)
p <
Total Brain Volume
46162 (642)
64958 (1341)
41% 0.0001
Cortex 32636 (736)
46779 (1224)
43% 0.0001
Diencephalon 5118 (85) 6892 (189) 35% 0.0001
Total Hippocampus
1066 (42) 1464 (58) 37% 0.0003
Cerebellum 3919 (69) 5380 (93) 37% 0.0001
Brainstem 3422 (62) 4443 (144) 30% 0.0001
Table A.1. Brain growth from 2- to 5-weeks of age. All volumes are mean (SE) in mm3. Three animals of each sex were used at 2- and 5-weeks of age.
140
Region Intraobserver Interobserver
Total Brain Volume
Median Agreement, %
94.7 98.1
COV, % 4.5 4.0
Cortex
Median Agreement, %
94.4 97.2
COV, % 6.4 5.2
Diencephalon
Median Agreement, %
95.6 98.0
COV, % 5.3 4.3
Cerebellum
Median Agreement, %
95.2 96.6
COV, % 3.0 2.0
Total Hippocampus
Median Agreement, %
94.8 98.8
COV, % 2.1 2.5
Brainstem
Median Agreement, %
94.9 97.4
COV, % 4.9 3.5
Table A.2. Intra- and interobserver agreements and coefficients of variation for segmented regions. All values are percentages.
141
A.7 References Beauchamp, M. H., D. K. Thompson, et al. (2008). "Preterm infant hippocampal volumes
correlate with later working memory deficits." Brain 131(Pt 11): 2986-2994. Bellinger, D. A., E. P. Merricks, et al. (2006). "Swine models of type 2 diabetes mellitus: insulin
resistance, glucose tolerance, and cardiovascular complications." ILAR J 47(3): 243-258. Björkman, S. T., S. M. Miller, et al. (2010). "Seizures are associated with brain injury severity in
a neonatal model of hypoxia-ischemia." Neuroscience 166(1): 157-167. Bland, J. M. and D. G. Altman (1986). "Statistical methods for assessing agreement between
two methods of clinical measurement." Lancet 1(8476): 307-310. Boardman, J. P., S. J. Counsell, et al. (2006). "Abnormal deep grey matter development
following preterm birth detected using deformation-based morphometry." Neuroimage 32(1): 70-78.
Bobinski, M., M. J. de Leon, et al. (2000). "The histological validation of post mortem magnetic
resonance imaging-determined hippocampal volume in Alzheimer's disease." Neuroscience 95(3): 721-725.
Bokde, A. L. W., S. J. Teipel, et al. (2005). "Reliable manual segmentation of the frontal,
parietal, temporal, and occipital lobes on magnetic resonance images of healthy subjects." Brain Res Brain Res Protoc 14(3): 135-145.
Brinkmann, B. H., A. Manduca, et al. (1998). "Optimized homomorphic unsharp masking for MR
grayscale inhomogeneity correction." IEEE Trans Med Imaging 17(2): 161-171. Cotter, D., K. Miszkiel, et al. (1999). "The assessment of postmortem brain volume; a
comparison of stereological and planimetric methodologies." Neuroradiology 41(7): 493-496.
Dickerson, J. W. T. and J. Dobbing (1967). "Prenatal and postnatal growth and development of
the central nervous system of the pig." Proc Biol Sci 166(1005): 384-395. Dilger, R. N. and R. W. Johnson (2010). "Behavioral assessment of cognitive function using a
translational neonatal piglet model." Brain Behav Immun 24(7): 1156-1165. Dixon, J. A. and F. G. Spinale (2009). "Large animal models of heart failure: a critical link in the
translation of basic science to clinical practice." Circ Heart Fail 2(3): 262-271. Dobbing, J. and J. Sands (1979). "Comparative aspects of the brain growth spurt." Early Hum
Dev 3(1): 79-83. Dubois, J., M. Benders, et al. (2008). "Mapping the early cortical folding process in the preterm
newborn brain." Cereb Cortex 18(6): 1444-1454. Felix, B., M. E. Leger, et al. (1999). "Stereotaxic atlas of the pig brain." Brain Res Bull 49(1-2):
1-137.
142
Ganessunker, D., H. R. Gaskins, et al. (1999). "Total parenteral nutrition alters molecular and cellular indices of intestinal inflammation in neonatal piglets." JPEN J Parenter Enteral Nutr 23(6): 337-344.
Giedd, J. N., J. Blumenthal, et al. (1999). "Brain development during childhood and
adolescence: a longitudinal MRI study." Nat Neurosci 2(10): 861-863. Giedd, J. N., A. C. Vaituzis, et al. (1996). "Quantitative MRI of the temporal lobe, amygdala, and
hippocampus in normal human development: ages 4-18 years." J Comp Neurol 366(2): 223-230.
Gilmore, J. H., W. Lin, et al. (2007). "Regional gray matter growth, sexual dimorphism, and
cerebral asymmetry in the neonatal brain." J Neurosci 27(6): 1255-1260. Institution, B. S. (1998). Accuracy (trueness and precision) of measurement methods and
results. Alternative methods for the determination of the precision of a standard measurement method. BS 5725, British Standards Institution. 5725-5.
Jelsing, J., E. Rostrup, et al. (2005). "Assessment of in vivo MR imaging compared to physical
sections in vitro--a quantitative study of brain volumes using stereology." Neuroimage 26(1): 57-65.
Knickmeyer, R. C., S. Gouttard, et al. (2008). "A structural MRI study of human brain
development from birth to 2 years." J Neurosci 28(47): 12176-12182. Kuluz, J., A. Samdani, et al. (2010). "Pediatric spinal cord injury in infant piglets: description of a
new large animal model and review of the literature." J Spinal Cord Med 33(1): 43-57. Lind, N. M., A. Moustgaard, et al. (2007). "The use of pigs in neuroscience: modeling brain
disorders." Neurosci Biobehav Rev 31(5): 728-751. Miller, E. R. and D. E. Ullrey (1987). "The pig as a model for human nutrition." Annu Rev Nutr 7:
361-382. Munkeby, B. H., C. De Lange, et al. (2008). "A piglet model for detection of hypoxic-ischemic
brain injury with magnetic resonance imaging." Acta Radiol 49(9): 1049-1057. Pfefferbaum, A., D. H. Mathalon, et al. (1994). "A quantitative magnetic resonance imaging
study of changes in brain morphology from infancy to late adulthood." Arch Neurol 51(9): 874-887.
Pond, W. G., S. L. Boleman, et al. (2000). "Perinatal ontogeny of brain growth in the domestic
pig." Proc Soc Exp Biol Med 223(1): 102-108. Rosendal, F., M. Pedersen, et al. (2009). "MRI protocol for in vivo visualization of the Gottingen
minipig brain improves targeting in experimental functional neurosurgery." Brain Res Bull 79(1): 41-45.
Saikali, S., P. Meurice, et al. (2010). "A three-dimensional digital segmented and deformable
brain atlas of the domestic pig." J Neurosci Meth 192(1): 102-109.
143
Sandberg, D. I., K. M. Crandall, et al. (2010). "Pharmacokinetic analysis of etoposide distribution after administration directly into the fourth ventricle in a piglet model." J Neurooncol 97(1): 25-32.
Shi, F., Y. Fan, et al. (2010). "Neonatal brain image segmentation in longitudinal MRI studies."
Neuroimage 49(1): 391-400. Thibault, K. L. and S. S. Margulies (1998). "Age-dependent material properties of the porcine
cerebrum: effect on pediatric inertial head injury criteria." J Biomech 31(12): 1119-1126. Thompson, D. K., S. J. Wood, et al. (2008). "Neonate hippocampal volumes: prematurity,
perinatal predictors, and 2-year outcome." Ann Neurol 63(5): 642-651. Wood, N. S., N. Marlow, et al. (2000). "Neurologic and developmental disability after extremely
preterm birth." N Engl J Med 343(6): 378-384. Yu, X., Y. Zhang, et al. (2010). "Comprehensive brain MRI segmentation in high risk preterm
newborns." PLoS One 5(11): e13874.
144
APPENDIX B
BRAIN GROWTH OF THE DOMESTIC PIG (SUS SCROFA) FROM 2 TO 24 WEEKS OF AGE: A LONGITUDINAL MRI STUDY
B.1 Abstract
An animal model with brain growth similar to humans, that can be used in MRI studies to
investigate brain development, would be valuable. Our lab has developed and validated MRI
methods for regional brain volume quantification in the neonatal piglet. The aim of this study
was to utilize the MRI-based volume quantification technique in a longitudinal study to
determine brain growth in domestic pigs from 2 to 24 wks of age. MRI data were acquired from
pigs 2 to 24 wks of age using a three-dimensional MPRAGE sequence on a MAGNETOM Trio
3T imager. Manual segmentation was performed for volume estimates of total brain, cortical,
diencephalon, brainstem, cerebellar, and hippocampal regions. Logistic modeling procedures
were used to characterize brain growth. Total brain volume increased 130% (± 12%) and 121%
(± 7%) from 2 to 24 wks in males and females, respectively. The maximum increase in total
brain volume occurred about age 4 wks and 95% of whole brain growth occurred by age 21-23
wks. Logistical modeling suggests there are sexually dimorphic effects on brain growth. For
example, in females, cortex was smaller (P = 0.04). Furthermore, the maximum growth of the
hippocampus occurred about 5 wks earlier in females than males, and the window for
hippocampal growth was significantly shorter in females than males (P = 0.02, P = 0.002
respectively) . These sexual dimorphisms are similar to what is seen in humans. In addition to
providing important data on brain growth for pigs, this study shows pigs can be used to obtain
longitudinal MRI data. The large increase in brain volume in the postnatal period is similar to
human neonates and suggests pigs can be used to investigate brain development.
This material has been previously published. The copyright owner has provided permission to reprint. Conrad MS, Dilger RN, Johnson RW (2012) Brain growth of the domestic pig (Sus scrofa) from 2 to 24 weeks of age: A longitudinal MRI study. Dev Neuroscience 34(4) 291-8
145
B.2 Introduction
At birth the human infant brain is just 25% of adult size. It undergoes massive postnatal
growth so by age 2 overall brain size reaches about 85% of adult volume (Knickmeyer, Gouttard
et al. 2008). This period of accelerated brain growth may be the most important phase of
postnatal brain development in humans, as it is the result of synaptogenesis, gliosis, and
myelination (Rice and Barone 2000). The rapid growth and development of the brain is
accompanied by an equally rapid development of cognitive and motor function (Forssberg 1999;
Rovee-Collier and Barr 2007). Thus, this early postnatal phase is recognized as a period of
increased vulnerability to injury, and disruption of neurodevelopment by environmental insults
may have long-lasting or permanent effects on brain structure and function. The idea that
environmental insults in an early sensitive period affect behavior later is supported by studies
showing that early-life stress increases the likelihood for several neuropsychological
disturbances later in life (Hackman, Farah et al. 2010); and that iron deficiency between age 6
months and 2 years leads to long-term deficits in learning and memory (Graham, Heim et al.
1999; McCann and Ames 2007; Thomas, Grant et al. 2009). Nonetheless, very little is known
about how human brain growth and development proceeds during this period let alone how
insults interfere.
Quantitative magnetic resonance imaging (MRI) has emerged as a powerful tool for
assessing human brain development. However, most large scale MRI studies of brain
development have been done with older children (>4 years of age); and use of MRI in infants
and younger children has mostly been restricted to those born preterm (Giedd, Snell et al. 1996;
Peterson, Anderson et al. 2003). Only recently, was MRI used to assess brain growth in healthy
full term infants from birth to age 2 years (Knickmeyer, Gouttard et al. 2008). This study has
provided an impressive view of the massive brain growth and development that occurs postnatal
in healthy full term infants and underscores this as being an influential period. Although this
application of MRI in human infants is impressive, the ability to address mechanistic questions
146
related to early-life insults cannot be readily done due to practical or ethical concerns.
Therefore, a pre-clinical animal model with brain growth and development similar to humans
that can be used in MRI studies to investigate how different insults at critical periods of rapid
brain growth affect development and function would be valuable.
Due to its anatomic and physiologic similarities with humans, the domestic pig (Sus
scrofa) is a preferred pre-clinical model in several areas including pediatric nutrition and organ
transplant surgery (Miller and Ullrey 1987; Dixon and Spinale 2009). Because pigs and humans
are thought to share similar brain growth and development patterns, appeal for the pig as a
model for human infant brain development has increased. The major brain growth spurt
extends from late prenatal to early postnatal in both the pig and human, which is different from
other common animal models (Dobbing and Sands 1979). Additionally, gross anatomical
features including gyral pattern and distribution of grey and white matter of the neonatal porcine
brain are similar to that of human infants (Dickerson and Dobbing 1967; Pond, Boleman et al.
2000; Lind, Moustgaard et al. 2007). The physical size of piglets also allows for quantitative
structural MRI of the brain using clinical scanners. Structural MRI scanning and analysis
protocols have been developed for neonatal and adult domestic pigs (Saikali, Meurice et al.
2010; Conrad, Dilger et al. 2012) and for adult Göttingen pigs (Jelsing, Rostrup et al. 2005).
More advanced techniques including, functional MRI and diffusion tensor imaging have also
been used (Fang, Lorke et al. 2005; Munkeby, De Lange et al. 2008). However, we are
unaware of previous studies where brain growth trajectory was assessed in pigs using MRI in a
longitudinal study. Previously, our lab has developed MRI methods for quantifying brain region
volumes in the neonatal piglet (Conrad, Dilger et al. 2012). The purpose here was to use these
techniques to determine total brain and brain region volumes in a cohort of male and female
domestic pigs on 7 occasions from age 2 wks to near sexual maturity at age 24 wks. The
results using this noninvasive longitudinal approach confirm substantial postnatal brain growth
in domestic pigs. Furthermore, using logistic modeling procedures to determine the maximum
147
volume and the age of maximum growth rate, we revealed a number of attributes of pig brain
growth that are similar to humans, including evidence for sexual dimorphic effects.
B.3 Material and Methods
Subjects
A total of fifteen pigs, six intact male and nine female (Sus scrofa domestica, York
breed), from seven litters, were obtained from the University of Illinois swine herd at two days of
age and placed into an artificial rearing system that was previously described (Dilger and
Johnson 2010). Briefly, each piglet was housed individually in an acrylic-sided caging unit with
a 12 hour light/dark cycle. Ambient temperature was maintained at 27˚C with intra-cage
temperatures maintained at 32˚C using overhead radiant heaters. Piglets were provided a
commercially available liquid milk replacer (Milk Specialties Global Animal Nutrition,
Carpentersville, IL) until 3 wks of age at which time they were moved to individual floor pens
(4.65 m2) and provided ad libitum access to water and a corn-soybean meal based diet
formulated to provide recommended levels of all essential nutrients (NRC 1998). Pigs were
weighed weekly to determine weight gain. All procedures were in accordance with the National
Institute of Health Guidelines for the Care and Use of Laboratory Animals and approved by the
University of Illinois Institutional Animal Care and Use Committee.
Image Acquisition
At 2 wks of age pigs were subjected to MRI scanning procedures to obtain brain images
for in vivo estimation of volume. Before scanning, pigs were anesthetized by intramuscular
injection of a telazol:ketamine:xylazine solution (TKX; 4.4 mg/kg body weight), and placed in a
prone position in the MRI scanner. Pigs remained anesthetized during the MRI scanning
procedure (approximately 20 min) and were returned to their pen afterwards where they were
monitored until recovering from anesthesia. All pigs underwent additional MRI scans at 4, 8, 12,
148
16, 20, and 24 wks of age. At the conclusion of the 24 wk MRI scan, pigs were euthanized by
intracardiac administration of an overdose of sodium pentobarbital (390 mg/ml Fatal Plus -
administered at 1 ml/5 kg body weight).
Magnetic resonance scanning was conducted using a three-dimensional T1-weighted
magnetization prepared gradient echo (MPRAGE) sequence. A 32-channel head coil was used
for pigs up to 8 wks of age, and a flexible 6-channel coil was used thereafter. The parameters
used were: repetition time, TR = 1900 ms; echo time, TE = 2.48 ms; inversion time, TI = 900
ms, flip angle = 9˚, matrix = 256 x 256 (interpolated to 512 x 512), slice thickness = 1.0 mm.
The final voxel size was 0.35 mm x 0.35 mm x 1.0 mm.
Brain Region Volume Estimation
Digital Imaging and Communications in Medicine (DICOM) images were imported and
analyzed using three-dimensional visualization software (AMIRA®, Visage Imaging, Inc., San
Diego, CA). The whole brain, cortex, diencephalon, cerebellum, hippocampus, and brainstem
were manually segmented using a Wacom Cintiq 21UX graphic input screen and pen (Wacom,
Vancouver, WA) and criteria previously described (Conrad, Dilger et al. 2012). Each region was
manually segmented in the three anatomical planes using the atlas by Felix et al. (Félix, Léger
et al. 1999), and regional volume estimates were calculated using the AMIRA© software. The
reliability of this technique in pigs has been validated (Conrad, Dilger et al. 2012).
Statistics
All data analysis was done using SAS software (SAS, Cary, NC). Weight gain data were
subjected to a repeated two-way mixed model ANOVA (Age × Sex). Brain region growth curves
149
were constructed for each brain area in each individual pig using the logistic growth model
shown below (Grossman and Koops 2003) and the proc nlin function of SAS:
BCagee
Avolumeregionbrain
/)(1
Parameter estimations were computed for maximum brain region volume (A), “duration”
of growth (B), and age of maximum growth rate (C) for each pig. The “duration” of growth term
(B) is defined as /3B where sigma is the standard deviation of the logistic function
(Gupta and Gnanadesikan 1966). Individual parameter estimations were then analyzed using a
mixed model including sex, litter, and the interaction. For analysis of hippocampal asymmetries,
hippocampal data were subjected to a two-way mixed model ANOVA (Sex × Hemisphere).
There were no significant differences due to hemisphere (data not shown); therefore left and
right hippocampal volumes were combined to create a total hippocampal volume measure. The
logistic growth model was further used to estimate the age when total brain and brain region
volume was 50%, 75%, and 95% of maximum volume. Significance level was set at p < 0.05.
B.4 Results
Body Weight
Whole body weight for male and female pigs from 2 to 24 wks of age is shown in Figure
1. There were differences in body weight due to age (F(6,78) = 702.88, p = <0.0001), sex
(F(1, 13) = 14.25, p = 0.0023), and an Age x Sex interaction (F(6,78) = 2.51, p = 0.0283), with
males weighing more than females beginning at 20 wks of age.
Total Brain and Brain Region Volumes
A representative set of DICOM images from an individual female pig across all ages is
shown in Figure 2a, and DICOM images in Figure 2b show the brain of a 12 wk old female pig in
three planes. Total brain volume and volume of cortex, hippocampus, diencephalon,
150
cerebellum, and brain stem were calculated after manual segmentation in three planes and the
changes from 2 to 24 wks of age are shown in Figure 3 as are the growth curves estimated by
logistic modeling. The estimations for maximum volume, “duration” of growth, and the age
when maximum growth rate occurred are presented for both males and females in Table 1.
Total brain volume increased 130% (± 12%) and 121% (± 7%) from 2 to 24 wks in males
and females, respectively. The maximum increase in total brain volume occurred when pigs
were about 4 wks old and 95% of whole brain growth occurred by 21-23 wks of age. Cortical
volumes changed similarly although the estimated maximum volume was greater for males than
females (F(1,2) = 21.21, P = 0.04). In 2 wk old piglets, hippocampal volume (right and left
hemispheres combined) was similar in males and females. However, the growth trajectory of
the hippocampus was sex dependent and at 24 wks, the hippocampus was larger in males than
females (F(1,2) = 141.60, P = 0.007). The maximum increase in hippocampal volume occurred
earlier in females than males (3 wks of age vs. 8 wks of age; F(1,2) = 51.68, P = 0.02). Based
on logistic modeling, 95% of hippocampal growth occurred in females by 24 wks of age whereas
in males this milestone was not expected until after 39 wks of age (F(1,2) = 551.99, P = 0.002).
In the diencephalon, there were no significant differences in maximum volume, duration of
growth, and age of maximum growth rate due to sex. For cerebellar growth, males showed a
significantly larger maximum volume than females (F(1,2) = 35.98, P = 0.03), but there were no
sex differences in duration of growth. In males the age of maximum growth rate occurred
roughly one wk later than females (F(1,2) = 40.14, P = 0.02). The maximum volume of the
brainstem was similar for males and females, but the age of maximum growth rate occurred
about a wk later in males (F(1,2) = 75.53, P = 0.01).
From the logistic growth model we further estimated the age when total brain and brain
region volumes were 50%, 75%, and 95% of maximum volumes (Table 2). For example, total
brain volume of males was estimated to be 50% and 75% of maximum volume at 4 and 11 wks
151
of age, respectively. On the other hand, volume of the hippocampus in males was estimated to
be 50% and 75% of maximum volume at about 8 and 20 wks of age, respectively.
B.5 Discussion
In the present study, each piglet was subjected to an MRI scan on seven occasions from
2 to 24 wks of age to determine brain growth throughout the neonatal period to near sexual
maturity. The results using this noninvasive longitudinal approach confirm substantial postnatal
brain growth, with the most rapid growth occurring at about age 4 wks when the brain is ~50%
of maximum volume. The period of rapid brain growth continued to about age 12 wks, when
growth rate began to slow. Results from two recent cross-sectional studies where MRI was
used to estimate brain volume of pigs at different ages and weights, also suggest rapid brain
growth in the postnatal period (Winter, Dorner et al. 2011; Conrad, Dilger et al. 2012). The large
increase in total brain volume in the postnatal period is similar to human neonates and suggests
this is a critical period where disruption of developmental processes by environmental insults
can have long-lasting or permanent effects on brain structure and function. Therefore, the
domestic pig may serve as a pre-clinical model for postnatal human brain development.
Quantitative MRI has yielded important information on brain development in early
childhood and adolescence (Giedd, Snell et al. 1996), but because of potential health concerns
for infants, few studies have focused on the period from birth to 4 years of age when dramatic
brain development occurs. Most MRI studies of infants have been with those born prematurely
or with significant health complications. Only recently was MRI used to assess structural brain
development of healthy full term infants from birth to 2 years of age (Knickmeyer, Gouttard et al.
2008). The first year after birth, total brain volume more than doubled and hemispheric grey and
white matter increased 149% and 11%, respectively. With total brain volume estimates for
healthy infants and healthy adults (Lieberman, Tollefson et al. 2005) it was determined that total
brain volume at 2-4 wks of age is ~36% of adult volume; and at 1 year and 2 years of age, total
152
brain volume is ~72% and 83% of adult volume, respectively (Knickmeyer, Gouttard et al.
2008). Thus, in healthy human infants there is enormous brain growth the first year after birth,
making this a period of increased vulnerability to injury. Investigating questions related to early-
life environmental insults and brain development and function is difficult in human infants due to
obvious practical and ethical concerns. Thus, in addition to providing important data on brain
growth trajectory for pigs, the present study shows pigs can be used to obtain longitudinal data
in MRI studies. This indicates pigs can be used in MRI studies to investigate how different
insults at critical periods of rapid brain growth affect development and function. This is
particularly useful because unlike some common animal models, piglets can be tested in
learning and memory tasks at an early age when brain growth is maximal (Dilger and Johnson
2010; Elmore, Dilger et al. 2012)
Despite substantial brain growth the first 2 years after birth, in humans total brain volume
continues to increase until about puberty, with the total brain volume peaking at 10.5 years of
age in females and 14.5 years of age in males (Lenroot, Gogtay et al. 2007); thereafter, total
brain volume diminishes with age so total cerebral volume follows an inverted U-shaped
trajectory (Giedd and Rapoport 2010). Maximum total brain volume in human males is 8-10%
larger than females (Goldstein, Seidman et al. 2001). In the present study, total brain volume
data obtained from pigs age 2 to 24 wks were best described using a logistic (sigmoidal) model.
Consistent with what has been reported for humans, the logistic model suggested maximum
total brain volume for male pigs was ~8% larger than for females. The final MRI scan occurred
near puberty and how total brain volume for pigs might change further into adulthood is not
known.
From the present study, we cannot determine composition of brain growth. In humans,
most neurogenesis and migration occurs early in the prenatal period (starting in the first month
of gestation) with astrocyte and oligigodendrocyte proliferation extending from the late prenatal
period into the postnatal period (Rice and Barone 2000). The increase in grey matter the first
153
few years of life is likely due to glial cell proliferation, dendritic and axonal arborization, and
synapse formation, while white matter increases are due to myelination (Mrzljak, Uylings et al.
1991; Miller and Ono 1998). Similar to humans, the domestic pig does not have substantial
neocortical neurogenesis in the postnatal period (Jelsing, Nielsen et al. 2006). This is different
than the Göttingen minipig, which has significant neuronal and glia cell development in the
postnatal period (Jelsing, Nielsen et al. 2006). Therefore, the substantial increase in brain
volume seen in the current study is likely due to changes in the neuropil (dendrites, axons, glia)
and myelination, but not neurogenesis. A limitation to our study is that grey and white matter
volumes could not be determined due to the use of manual segmentation. However, a recent
cross sectional study of domestic pigs suggests grey and white matter volume continues to
increase at least through the first 12 wks of age (Winter, Dorner et al. 2011).
In humans, the cerebellum is 8-13% larger in males than females (Giedd, Snell et al.
1996; Tiemeier, Lenroot et al. 2010). Here we show that the cerebellum in adult male pigs is
roughly 10% larger than in females. Also in humans, males show peak cerebellar volume later
in life at 15.6 years of age compared to 11.8 in females (Tiemeier, Lenroot et al. 2010). We did
not find any difference in the age range of cerebellar growth in pigs, but we did find that the
peak growth rate occurred earlier in life for females. Developmental growth patterns of
subcortical structures depend on the age range being studied. During early postnatal
development, newborn through 2 years of age, subcortical structures develop similar in males
and females (Knickmeyer, Gouttard et al. 2008). Later in development, from 5 to 18 years of
age, subcortical structures including the putamen and globus pallidus have been shown to be
larger in males (Giedd, Snell et al. 1996). In this study, we found that the diencephalon and
brainstem have similar growth patterns in male and female pigs.
Studies on human hippocampal development have shown significant growth from birth
through 2 years of age (Pfluger, Weil et al. 1999). At age 2 years, the hippocampus is about
85% of adult size but data for males and females have not been reported separately so if sexual
154
dimorphic effects are present early on is not clear. Sexual dimorphic effects are, however,
present later. In females, e.g., maximum volume is smaller and achieved at a younger age
compared to males (Giedd, Vaituzis et al. 1996; Pfluger, Weil et al. 1999). Our data suggest a
similar sex-dependent growth trajectory for the pig hippocampus. For example, the period of
maximum growth occurred about 5 wks earlier in females than males, and the window for
hippocampal growth was significantly shorter in females than males. One feature of
hippocampal development that is different between pigs and humans relates to left/right
asymmetries. Multiple studies of human subjects have determined that the right hippocampus
is significantly larger in both males and females, and this difference is present from birth (Giedd,
Vaituzis et al. 1996; Pfluger, Weil et al. 1999; Utsunomiya, Takano et al. 1999; Thompson,
Wood et al. 2009). In our study, no difference was found between the left and right
hippocampal formation. The importance of hippocampal asymmetry is unknown, but
asymmetries can be affected by environmental insults, and changes in asymmetries have been
seen in patients with psychiatric illness (Stefanis, Frangou et al. 1999; Wang, Joshi et al. 2001).
Although we did not find differences, it may be an important measure to include in future
studies.
The present study has several other limitations. First, the study included a small sample
size, meaning the sexual dimorphisms in brain growth and development observed should be
cautiously interpreted. However, our confidence is bolstered because variation was minimized
by: (1) using a longitudinal design where each pig served as its own control, (2) including
littermates of each sex when possible; and (3) maintaining pigs individually in an identical
environment throughout the study. A second limitation was the use of manual segmentation for
quantifying brain region volume. Technical challenges arise when imaging the neonatal brain
including poor spatial resolution and low tissue contrast (Shi, Fan et al. 2010). In addition, there
is no MRI-based atlas available for the neonatal piglet. Construction of an atlas set across
multiple ages in the pig would allow for more complex automated segmentation protocols.
155
Nonetheless, this manual segmentation protocol has been used previously and is a reliable
method for determining brain region volume changes in the pig (Conrad, Dilger et al. 2012). A
third limitation is that pigs were only scanned through 24 wks of age, which is near sexual
maturity (Lind, Moustgaard et al. 2007). Beyond this age, most domestic pigs are too large to fit
in a standard MRI machine bore. This limitation precluded us from extending the study. Thus,
use of an open bore magnet may be useful to track changes past this age in order to determine
if there are brain volume loses after puberty as seen in many brain areas in the human (Lenroot,
Gogtay et al. 2007). Fourth, the data in this study suggests there are a few differences in brain
growth in pigs compared to humans. For example, we did not find any sexual dimorphisms in
maximum volumes of the diencephalon and brainstem regions; nor did we observe asymmetries
in the hippocampus. These differences present a limitation of the pig as an animal model for
human neurodevelopment.
In summary, the present study shows the normal brain growth pattern in the domestic
pig from 2 to 24 wks of age. The brain undergoes significant growth during this period and
several sexual dimorphisms affecting growth, including maximum volume, appear to be similar
for humans and pigs. The study shows that domestic pigs can be used in longitudinal MRI-
based studies to investigate brain growth and development, and suggests the pig can serve as
a preclinical model for human neurodevelopment.
156
B.6 Figures and Tables
Figure B.1. Body weight of pigs from 2 to 24 wks of age. Body weight of male pigs higher than female pigs starting at 20 wks of age. The error bars indicate SEM. * = p < 0.01 ** = p < 0.001
157
Figure B.2. Representative MRI images displaying brain growth from 2 to 24 wks of age. The top images are T1-weighted axial sections (skull stripped) from a female pig at each age (A). The bottom images show the coronal, axial, and sagittal views left to right of a 12 wk old female pig (B). Segmentation is shown for cortex (white), diencephalon (green), hippocampi (blue), cerebellum (red), and brainstem (yellow).
158
Figure B.3. Brain region growth curves and logistic models. The graphs show total brain and brain region volumes for male and female pigs from 2 to 24 wks of age, as well as logistic growth models for male and female pigs. Error bars indicated SEM.
159
Table B.1. Average logistic growth parameter estimates by sex for each brain region. The averages by sex for maximum volume (A), “duration” of growth (B), and age of maximum growth rate (C) are indicated. Mixed model analysis was used to determine the main effect of sex on each parameter.
160
Percent of Adult Volume 50% 75% 95%
Age(weeks)
SEMAge
(weeks)SEM
Age(weeks)
SEM
Total Brain
Volume Male 4.08 0.12 11.12 0.30 22.94 0.61
Female 3.81 0.14 10.30 0.30 21.19 0.57
Cortex Male 4.01 0.22 11.34 0.53 23.66 1.07
Female 3.70 0.19 10.22 0.39 21.19 0.74
Whole Hippocampus
Male 8.27 1.24 19.82 2.37 39.21 4.36
Female 3.27 0.16 11.35 0.25 24.92 0.43
Diencephalon Male 4.36 0.22 10.74 0.45 21.47 0.87
Female 4.42 0.07 11.26 0.12 22.75 0.26
Cerebellum Male 5.54 0.16 11.97 0.43 22.76 0.89
Female 4.89 0.05 10.68 0.13 20.41 0.29
Brainstem Male 5.95 0.46 14.44 0.92 28.69 1.72
Female 4.61 0.18 11.92 0.46 24.20 0.93
Table B.2. Age when the brain volumes reach 50, 75, and 95% of maximum volume. Using the logistic models, estimates, with standard errors, were computed for the ages when different brain regions of each sex reach 50, 75, and 95% of adult brain volume for each measure. These data can be compared to human data to determine comparable growth timelines for designing experimental time windows.
161
B.7 References Conrad, M. S., R. N. Dilger, et al. (2012). "Magnetic resonance imaging of the neonatal piglet
brain." Pediatr Res 71(2): 179-184. Dickerson, J. W. T. and J. Dobbing (1967). "Prenatal and postnatal growth and development of
the central nervous system of the pig." Proc Biol Sci 166(1005): 384-395. Dilger, R. N. and R. W. Johnson (2010). "Behavioral assessment of cognitive function using a
translational neonatal piglet model." Brain Behav Immun 24(7): 1156-1165. Dixon, J. A. and F. G. Spinale (2009). "Large animal models of heart failure: a critical link in the
translation of basic science to clinical practice." Circ Heart Fail 2(3): 262-271. Dobbing, J. and J. Sands (1979). "Comparative aspects of the brain growth spurt." Early Human
Development 3(1): 79-83. Elmore, M. R., R. N. Dilger, et al. (2012). "Place and direction learning in a spatial T-maze task
by neonatal piglets." Anim Cogn 15(4): 667-676. Fang, M., D. E. Lorke, et al. (2005). "Postnatal changes in functional activities of the pig's brain:
a combined functional magnetic resonance imaging and immunohistochemical study." Neurosignals 14(5): 222-233.
Félix, B., M.-E. Léger, et al. (1999). "Stereotaxic atlas of the pig brain." Brain Res Bull 49(1-2):
1-137. Forssberg, H. (1999). "Neural control of human motor development." Curr Opin Neurobiol 9(6):
676-682. Giedd, J. N. and J. L. Rapoport (2010). "Structural MRI of Pediatric Brain Development: What
Have We Learned and Where Are We Going?" Neuron 67(5): 728-734. Giedd, J. N., J. W. Snell, et al. (1996). "Quantitative magnetic resonance imaging of human
brain development: ages 4-18." Cereb Cortex 6(4): 551-560. Giedd, J. N., A. C. Vaituzis, et al. (1996). "Quantitative MRI of the temporal lobe, amygdala, and
hippocampus in normal human development: ages 4-18 years." J Comp Neurol 366(2): 223-230.
Goldstein, J. M., L. J. Seidman, et al. (2001). "Normal sexual dimorphism of the adult human
brain assessed by in vivo magnetic resonance imaging." Cereb Cortex 11(6): 490-497. Graham, Y. P., C. Heim, et al. (1999). "The effects of neonatal stress on brain development:
Implications for psychopathology." Dev Psychopathol 11(03): 545-565. Grossman, M. and W. J. Koops (2003). "Modeling extended lactation curves of dairy cattle: a
biological basis for the multiphasic approach." J Dairy Sci 86(3): 988-998.
162
Gupta, S. S. and M. Gnanadesikan (1966). "Estimation of the parameters of the logistic distribution." Biometrika 53(3-4): 565-570.
Hackman, D. A., M. J. Farah, et al. (2010). "Socioeconomic status and the brain: mechanistic
insights from human and animal research." Nat Rev Neurosci 11(9): 651-659. Jelsing, J., R. Nielsen, et al. (2006). "The postnatal development of neocortical neurons and
glial cells in the Gottingen minipig and the domestic pig brain." J Exp Biol 209(Pt 8): 1454-1462.
Jelsing, J., E. Rostrup, et al. (2005). "Assessment of in vivo MR imaging compared to physical
sections in vitro--a quantitative study of brain volumes using stereology." Neuroimage 26(1): 57-65.
Knickmeyer, R. C., S. Gouttard, et al. (2008). "A structural MRI study of human brain
development from birth to 2 years." J Neurosci 28(47): 12176-12182. Lenroot, R. K., N. Gogtay, et al. (2007). "Sexual dimorphism of brain developmental trajectories
during childhood and adolescence." NeuroImage 36(4): 1065-1073. Lieberman, J. A., G. D. Tollefson, et al. (2005). "Antipsychotic Drug Effects on Brain Morphology
in First-Episode Psychosis." Arch Gen Psychiatry 62(4): 361-370. Lind, N. M., A. Moustgaard, et al. (2007). "The use of pigs in neuroscience: modeling brain
disorders." Neurosci Biobehav Rev 31(5): 728-751. McCann, J. C. and B. N. Ames (2007). "An overview of evidence for a causal relation between
iron deficiency during development and deficits in cognitive or behavioral function." Am J Clin Nutr 85(4): 931-945.
Miller, E. R. and D. E. Ullrey (1987). "The pig as a model for human nutrition." Annu Rev Nutr 7:
361-382. Miller, R. H. and K. Ono (1998). "Morphological analysis of the early stages of oligodendrocyte
development in the vertebrate central nervous system." Microsc Res Techniq 41(5): 441-453.
Mrzljak, L., H. B. M. Uylings, et al. (1991). Chapter 9 Neuronal development in human prefrontal
cortex in prenatal and postnatal stages. Progress in Brain Research. C. G. V. E. J. P. C. D. B. M. A. C. H.B.M. Uylings and M. G. P. Feenstra, Elsevier. Volume 85: 185-222.
Munkeby, B. H., C. De Lange, et al. (2008). "A piglet model for detection of hypoxic-ischemic
brain injury with magnetic resonance imaging." Acta Radiol 49(9): 1049-1057. NRC (1998). Nutrient Requirements of Swine. Washington, DC, National Academic Press. Peterson, B. S., A. W. Anderson, et al. (2003). "Regional Brain Volumes and Their Later
Neurodevelopmental Correlates in Term and Preterm Infants." Pediatrics 111(5): 939-948.
163
Pfluger, T., S. Weil, et al. (1999). "Normative volumetric data of the developing hippocampus in children based on magnetic resonance imaging." Epilepsia 40(4): 414-423.
Pond, W. G., S. L. Boleman, et al. (2000). "Perinatal ontogeny of brain growth in the domestic
pig." Proc Soc Exp Biol Med 223(1): 102-108. Rice, D. and S. Barone, Jr. (2000). "Critical periods of vulnerability for the developing nervous
system: evidence from humans and animal models." Environ Health Perspect 108 Suppl 3: 511-533.
Rovee-Collier, C. and R. Barr (2007). Infant Learning and Memory. Blackwell Handbook of
Infant Development, Blackwell Publishing Ltd: 139-168. Saikali, S., P. Meurice, et al. (2010). "A three-dimensional digital segmented and deformable
brain atlas of the domestic pig." J Neurosci Meth 192(1): 102-109. Shi, F., Y. Fan, et al. (2010). "Neonatal brain image segmentation in longitudinal MRI studies."
Neuroimage 49(1): 391-400. Stefanis, N., S. Frangou, et al. (1999). "Hippocampal volume reduction in schizophrenia: effects
of genetic risk and pregnancy and birth complications." Biological Psychiatry 46(5): 697-702.
Thomas, D. G., S. L. Grant, et al. (2009). "The role of iron in neurocognitive development." Dev
Neuropsychol 34(2): 196-222. Thompson, D. K., S. J. Wood, et al. (2009). "MR-determined hippocampal asymmetry in full-
term and preterm neonates." Hippocampus 19(2): 118-123. Tiemeier, H., R. K. Lenroot, et al. (2010). "Cerebellum development during childhood and
adolescence: A longitudinal morphometric MRI study." NeuroImage 49(1): 63-70. Utsunomiya, H., K. Takano, et al. (1999). "Development of the temporal lobe in infants and
children: analysis by MR-based volumetry." AJNR Am J Neuroradiol 20(4): 717-723. Wang, L., S. C. Joshi, et al. (2001). "Statistical Analysis of Hippocampal Asymmetry in
Schizophrenia." NeuroImage 14(3): 531-545. Winter, J. D., S. Dorner, et al. (2011). "Noninvasive MRI measures of microstructural and
cerebrovascular changes during normal swine brain development." Pediatr Res 69(5 Pt 1): 418-424.
164
APPENDIX C
AN IN VIVO THREE-DIMENSIONAL MAGNETIC RESONANCE IMAGING-BASED AVERAGED BRAIN AND ATLAS OF THE NEONATAL PIGLET (SUS SCROFA)
C.1 Abstract
Due to the fact that morphology and perinatal growth of the piglet brain is similar
to humans, use of the piglet as a translational animal model for neurodevelopmental
studies is increasing. Magnetic resonance imaging (MRI) can be a powerful tool to
study neurodevelopment in piglets, but many of the MRI resources have been produced
for adult humans. Here, we present an average in vivo MRI-based atlas specific for the
4-week-old piglet. In addition, we have developed probabilistic tissue classification
maps. These tools can be used with brain mapping software packages (e.g. SPM and
FSL) to aid in voxel-based morphometry and image analysis techniques. The atlas
enables efficient study of neurodevelopment in a highly tractable translational animal
with brain growth and development similar to humans.
165
C.2 Introduction
Use of the domestic pig (Sus scrofa) as a translational animal model for
neuroscience research is increasing (Lind, Moustgaard et al. 2007). The pig is an
attractive model because like humans, the major brain growth spurt extends from the
late prenatal to the postnatal period (Dobbing and Sands 1979). Gross anatomical
features, including gyral pattern and distribution of gray and white matter of the neonatal
piglet brain are similar to that of human infants (Dickerson and Dobbing 1967; Thibault
and Margulies 1998). Moreover, their physical size allows neuroimaging instruments
designed for humans to be used with piglets. Indeed, structural magnetic resonance
imaging (MRI), functional MRI, and positron emission tomography have all been
conducted in pigs (Ishizu, Smith et al. 2000; Watanabe, Andersen et al. 2001; Fang,
Lorke et al. 2005; Jakobsen, Pedersen et al. 2006). Finally, due to their precocial
nature, piglets can be weaned at birth or after caesarian delivery, maintained with
relative ease, and used in behavioral testing paradigms to assess learning at an early
age (Dilger and Johnson 2010; Elmore, Dilger et al. 2012). Thus, piglets represent a
gyrencephalic species with brain growth similar to humans that can be used in highly
controlled experiments to explore how environmental insults such as nutrient
deficiencies, infection, or social stress affect brain structure and function.
Previously, we reported MRI techniques for quantifying brain region volumes in
the neonatal piglet (Conrad, Dilger et al. 2012). These techniques were used to
quantify normal brain growth of the domestic pig from 2- to 24-weeks of age (Conrad,
Dilger et al. 2012). During this period, brain volume increased 121-130% and several
sexual dimorphisms were identified. For example, the maximum growth rate of the
166
hippocampus occurred 5 weeks earlier in females than males but the adult hippocampal
volume was greater in males, similar to patterns observed in humans (Giedd, Vaituzis
et al. 1996; Pfluger, Weil et al. 1999). In that study, labor-intensive manual
segmentation was used to estimate whole brain and brain region volume.
More advanced semi- and fully-automated structural analysis techniques, such
as voxel-based morphometry, have been used in human studies for over a decade
(Ashburner and Friston 2000). These techniques have also been used in rodents and
non-human primates (Sawiak, Wood et al. 2009; McLaren, Kosmatka et al. 2010). A
prerequisite to using these advanced methods in piglets is the availability of a
standardized atlas that serves as a template for registration and spatial normalization.
An example of a standardized template for humans is the MNI152 atlas (Evans, Collins
et al. 1993; Collins, Neelin et al. 1994). There are also atlases available for rodents and
non-human primates (Ma, Hof et al. 2005; McLaren, Kosmatka et al. 2009). A three-
dimensional digital brain atlas has been made for a 6-month old domestic pig using
high-resolution MRI and histological slices (Saikali, Meurice et al. 2010). Although this
is a very good atlas for localizing specific brain areas in the adult pig, it is not
appropriate for the young piglet because of the significant difference in brain size
(Conrad, Dilger et al. 2012). An atlas has also been created for the adult Göttingen
minipig, but due to size differences between breeds (15-18 kg for an adult Göttingen
minipig vs. 5-7 kg for a 4-week-old domestic pig), it also is not appropriate for use with
domestic neonatal piglets (Watanabe, Andersen et al. 2001). In addition, the Göttingen
minipig has significant neuronal and glial cell development in the postnatal period,
167
something not seen in humans and the domestic pig (Jelsing, Nielsen et al. 2006). This
suggests the domestic pig may be a better model for human neurodevelopment.
Therefore, our goal was to create an in vivo MRI-based atlas specifically for the 4-week-
old domestic piglet.
Here, we report the creation of a T1-weighted population-averaged template for
4-week-old piglets. Probabilistic tissue classification maps for gray matter, white matter,
and cerebrospinal fluid were also generated. Using the average template, we manually
segmented nineteen regions of interest to create the neonatal piglet atlas. These
packages, which are now publically available (http://pigmri.illinois.edu), will allow for
more sophisticated brain structure analysis, including voxel-based morphometry, in the
neonatal piglet.
C.3 Materials and Methods
Ethics Statement
All animal experiments were in accordance with the National Institute of Health
Guidelines for the Care and Use of Laboratory Animals and approved by the University
of Illinois at Urbana-Champaign Institutional Animal Care and Use Committee.
Subjects
Fifteen pigs, nine female and six male (Sus scrofa domestica, York breed), were
obtained from the University of Illinois swine heard. The pigs were placed into an
artificial rearing system 48-hours after birth (previously described by Dilger and Johnson
(Dilger and Johnson 2010)). Briefly, each pig was placed in an acrylic-sided cage (0.76
168
mL x 0.58 mW x 0.47 mH) and provided an enrichment toy and towel. Overhead
radiant heaters maintained the temperature at 27°C. Piglets were maintained on a 12-
hour light/dark cycle. Piglets were fed a commercially available piglet milk replacer
(Advance Liqui-wean, Milk Specialties Co., Dundee, IL) at 285 mL/kg body weight. The
piglets received this milk in 18 meals during the day using an automated liquid feeding
system (Dilger and Johnson 2010). The average body weight of the pigs at 2 and 28
days of age was 1.67 (± 0.08) kg and 5.35 (± 0.39) kg, respectively. The animals were
part of a longitudinal study characterizing the normal brain development of the pig
(Conrad, Dilger et al. 2012).
Image Acquisition
The image acquisition procedures have been described previously (Conrad,
Dilger et al. 2012). Briefly, piglets were transported to the MRI facility where they were
anesthetized with an intramuscular injection of a telazol:ketamine:xylazine (TKX; 4.4
mg/kg body weight), and placed in a prone position in a Siemens Trio 3 T imager
(Siemens, Erlangen, Germany). A three-dimensional T1-weighted magnetization
prepared gradient echo (MPRAGE) sequence was used with a 32-channel coil
(Siemens 32-channel head coil). The sequence parameters were: repetition time, TR =
1900 ms; echo time, TE = 2.48 ms; inversion time, TI = 900 ms, flip angle = 9°, matrix =
256 x 256 (interpolated to 512 x 512), slice thickness 1.00 mm; this produced a voxel
size of 0.35 mm x 0.35 mm x 1.0 mm. A total of 192 slices were acquired. The images
used for creating the atlas were acquired when pigs were 4-weeks of age (average
body weight, 5.35 kg ±0.39).
169
Averaged Brain Creation
First, the Digital Imaging and Communication in Medicine (DICOM) images were
reconstructed into 3D volumes for each pig. Using the 3D volumes, a binary mask was
manually drawn over the brain tissue for each pig using the FSLView function of FSL
(Analysis Group, FMRIB, Oxford, UK) (Jenkinson, Beckmann et al. 2012). The mask
was then used to extract the brain from the original data set. Next, the “fast” function of
FSL was used to create segmentation maps while also bias correcting for field
inhomogeneity and intensity normalization. An extracted brain from one pig was chosen
as the “template” data set and the other fourteen data sets were linearly registered with
a rigid-body transformation using the coregister module of SPM (University College
London, London, UK). The aligned images were averaged using “SPM Image
Calculator” to create the “averaged template.” An additional iteration was conducted to
linearly realign all of the original data sets to the averaged template to improve
alignment.
This template represents the average of all of the data sets, but does not factor in
the average shape or morphology of the brain. To compensate for this, we non-linearly
registered all of the original brains using the Normalization module in SPM to this
template to generate individual deformation fields for each animal. Many atlases,
including human infant atlases, use different variations of non-linear registration to
improve the averaged brain (Altaye, Holland et al. 2008; Shi, Fan et al. 2010). The
deformation fields for all pigs were then averaged. The inverse of this average
deformation field was then applied to the template in order to compensate for the
170
average brain shape. This resulted in an averaged brain with compensation for average
brain shape.
Tissue Probability Maps
The “fast” function in FSL does not require prior tissue maps as it uses a hidden
Markov random field model and an expectation-maximization algorithm (Zhang, Brady
et al. 2001). Using the “fast” function, we determined the partial volume tissue
estimation for each individual brain which had been linearly registered during the
second iteration above. The partial volume estimates were used as they approximate
the proportion of each tissue type in each voxel. Individual gray matter, white matter,
and CSF partial volume estimates (PVE) maps were created for each pig. The
deformation fields created for each pig during the normalization procedure during atlas
construction were then applied to the PVEs to bring them into the averaged brain space.
The PVE were then averaged by tissue type to create the prior probability maps. These
maps are provided as raw files and smoothed with a 1 mm full width at half maximum
(FWHM) Gaussian smoothing kernel (Evans, Kamber et al. 1994).
Manual Segmentation of Brain Regions of Interest
After completion of the averaged brain and tissue probability maps, both were
reoriented along the anterior and posterior commissure (ac-pc) and cropped to reduce
file size. The origin (zero reference point) was set to be the anterior limit of the posterior
commissure in the midsagittal plane to be consistent with previously published MRI and
histological atlases on adult pigs (Felix, Leger et al. 1999; Watanabe, Andersen et al.
171
2001; Saikali, Meurice et al. 2010). The origin can be seen in Figure 1. Using the
histological atlas by Felix et al. (Felix, Leger et al. 1999) as reference, nineteen regions
of interest (ROI) were identified and manually segmented on the averaged brain using
FSLView. Since no histological atlas is available for the piglet, these regions are larger,
generalized regions that could be easily identified and segmented on the averaged
brain. The ROIs were segmented in all three orthogonal views to increase accuracy. A
three-dimensional reconstruction of the regions of interest was created using AMIRA®
(Visage Imaging, Inc., San Diego, CA.)
C.4 Results and Discussion
Averaged Brain and Tissue Probability Maps
Our objective was to develop a T1-weighted average brain specific for the
neonatal piglet. This atlas was designed with features including high spatial resolution,
a standard coordinate space, easy access and visualization, and the ability to be linked
to current and future atlases, which are necessary for proper brain atlas construction
(Van Essen and Dierker 2007; McLaren, Kosmatka et al. 2009; Saikali, Meurice et al.
2010). Using MRI data from fifteen animals, a series of linear and non-linear
transformations were used to create the averaged brain. Axial slices through an
individual case and the averaged brain are shown in Figure 2. In addition, tissue
probability maps were created for gray matter, white matter and cerebrospinal fluid. A
coronal comparison of the three tissue types is shown in Figure 3. The tissue
probability maps give the probability that a voxel is classified as one of the three tissue
types.
172
Brain Regions
Using the neonatal piglet averaged brain, we manually segmented nineteen
regions. A list of these structures can be found in Table 1. These regions are provided
in a single NIfTI file containing corresponding region labels, as well as individual binary
files. Figure 4 shows a cross section of the brain highlighting a few regions of interest.
The combined image file can be deformed to new data sets to allow for region
identification using automated procedures. In addition, the individual region of interest
masks can be deformed to allow for quantitative volume estimation for each brain
region.
Applications
The goal of this study was to create an averaged brain for the neonatal piglet that
will allow for more sophisticated analysis including voxel-based morphometry. The
averaged brain serves as a starting template for voxel-based morphometry and can be
used with any software that uses the NIfTI file type. In addition, the averaged brain can
serve as a standard template for multimodal registration, including diffusion tensor
imaging. In data sets including diffusion tensor imaging, the atlas can also serve to
create predefined regions of interest for seeding or interrogation of imaging metrics.
Tissue probability maps serve as a priori inputs for tissue class segmentation.
We used the “fast” function from FSL to create tissue classifications for individual piglets
as this does not require priors. The resultant probability maps generated can be used
with the “Segment” function of SPM, as this software uses the priors in its segmentation
algorithm, and FSL which can also incorporate priors if they are available (Ashburner
and Friston 2000; Ashburner and Friston 2005).
173
Atlas Availability
The T1-weighted averaged brain, tissue probability maps, and ROI labels are freely
available at http://pigmri.illinois.edu. The files are freely distributed and fall under the
Illinois Open Source License.
174
C.5 Figures and Tables
Figure C.1. Location of origin for the atlas. Sagittal, coronal, and axial views of the origin location for the atlas.
175
Figure C.2. Comparison of an individual MRI data set and the atlas. The axial slices are from -18 mm to 18 mm in 6 mm increments. The top row is a representative individual case and the bottom row is the atlas.
176
Figure C.3. Tissue probability maps. Coronal slices through the origin showing the atlas T1 image, gray matter, white matter, and cerebrospinal fluid (CSF) tissue probability maps. The maps have been smoothed with a 1 mm FWHM Gaussian filter.
177
Figure C.4. Regions of interest. Sagittal, coronal, and axial slices through the origin showing manually segmented regions of interest.
178
Label Number Brain Region 1 Caudate 2 Cerebellum 3 Left Cortex 4 Right Cortex 5 Lateral Ventricle 6 Third Ventricle 7 Cerebral Aqueduct 8 Fourth Ventricle 9 Left Hippocampus 10 Right Hippocampus 11 Medulla 12 Midbrain 13 Pons 14 Putamen and Globus Pallidus 15 Hypothalamus 16 Thalamus 17 Olfactory Bulb 18 Corpus Callosum 19 Internal Capsule
Table C.1. The regions of interest found in the atlas and their respective label numbers.
179
C.6 References
Altaye, M., S. K. Holland, et al. (2008). "Infant brain probability templates for MRI segmentation and normalization." NeuroImage 43(4): 721-730.
Ashburner, J. and K. J. Friston (2000). "Voxel-Based Morphometry—The Methods."
NeuroImage 11(6): 805-821. Ashburner, J. and K. J. Friston (2005). "Unified segmentation." NeuroImage 26(3): 839-851. Collins, D. L., P. Neelin, et al. (1994). "Automatic 3D intersubject registration of MR volumetric
data in standardized Talairach space." J Comput Assist Tomogr 18(2): 192-205. Conrad, M. S., R. N. Dilger, et al. (2012). "Brain Growth of the Domestic Pig (Sus scrofa) from 2
to 24 Weeks of Age: A Longitudinal MRI Study." Dev Neurosci 34(4): 291-298. Conrad, M. S., R. N. Dilger, et al. (2012). "Magnetic resonance imaging of the neonatal piglet
brain." Pediatr Res 71(2): 179-184. Dickerson, J. W. T. and J. Dobbing (1967). "Prenatal and postnatal growth and development of
the central nervous system of the pig." Proc Biol Sci 166(1005): 384-395. Dilger, R. N. and R. W. Johnson (2010). "Behavioral assessment of cognitive function using a
translational neonatal piglet model." Brain Behav Immun 24(7): 1156-1165. Dobbing, J. and J. Sands (1979). "Comparative aspects of the brain growth spurt." Early Hum
Dev 3(1): 79-83. Elmore, M. R., R. N. Dilger, et al. (2012). "Place and direction learning in a spatial T-maze task
by neonatal piglets." Anim Cogn 15(4): 667-676. Evans, A. C., D. L. Collins, et al. (1993). 3D statistical neuroanatomical models from 305 MRI
volumes. Nuclear Science Symposium and Medical Imaging Conference, 1993., 1993 IEEE Conference Record.
Evans, A. C., M. Kamber, et al. (1994). An MRI-Based Probabilistic Atlas of Neuroanatomy.
Magnetic Resonance Scanning and Epilepsy. S. D. Shorvon, D. R. Fish, F. Andermann, G. M. Bydder and H. Stefan, Springer US. 264: 263-274.
Fang, M., D. E. Lorke, et al. (2005). "Postnatal changes in functional activities of the pig's brain:
a combined functional magnetic resonance imaging and immunohistochemical study." Neurosignals 14(5): 222-233.
Felix, B., M. E. Leger, et al. (1999). "Stereotaxic atlas of the pig brain." Brain Res Bull 49(1-2):
1-137. Giedd, J. N., A. C. Vaituzis, et al. (1996). "Quantitative MRI of the temporal lobe, amygdala, and
hippocampus in normal human development: ages 4-18 years." J Comp Neurol 366(2): 223-230.
180
Ishizu, K., D. F. Smith, et al. (2000). "Positron emission tomography of radioligand binding in porcine striatum in vivo: Haloperidol inhibition linked to endogenous ligand release." Synapse 38(1): 87-101.
Jakobsen, S., K. Pedersen, et al. (2006). "Detection of α2-Adrenergic Receptors in Brain of
Living Pig with 11C-Yohimbine." J Nucl Med 47(12): 2008-2015. Jelsing, J., R. Nielsen, et al. (2006). "The postnatal development of neocortical neurons and
glial cells in the Gottingen minipig and the domestic pig brain." J Exp Biol 209(Pt 8): 1454-1462.
Jenkinson, M., C. F. Beckmann, et al. (2012). "FSL." NeuroImage 62(2): 782-790. Lind, N. M., A. Moustgaard, et al. (2007). "The use of pigs in neuroscience: modeling brain
disorders." Neurosci Biobehav Rev 31(5): 728-751. Ma, Y., P. R. Hof, et al. (2005). "A three-dimensional digital atlas database of the adult
C57BL/6J mouse brain by magnetic resonance microscopy." Neuroscience 135(4): 1203-1215.
McLaren, D. G., K. J. Kosmatka, et al. (2010). "Rhesus macaque brain morphometry: a
methodological comparison of voxel-wise approaches." Methods 50(3): 157-165. McLaren, D. G., K. J. Kosmatka, et al. (2009). "A population-average MRI-based atlas collection
of the rhesus macaque." NeuroImage 45(1): 52-59. Pfluger, T., S. Weil, et al. (1999). "Normative volumetric data of the developing hippocampus in
children based on magnetic resonance imaging." Epilepsia 40(4): 414-423. Saikali, S., P. Meurice, et al. (2010). "A three-dimensional digital segmented and deformable
brain atlas of the domestic pig." J Neurosci Meth 192(1): 102-109. Sawiak, S. J., N. I. Wood, et al. (2009). "Voxel-based morphometry in the R6/2 transgenic
mouse reveals differences between genotypes not seen with manual 2D morphometry." Neurobiol Dis 33(1): 20-27.
Shi, F., Y. Fan, et al. (2010). "Neonatal brain image segmentation in longitudinal MRI studies."
Neuroimage 49(1): 391-400. Thibault, K. L. and S. S. Margulies (1998). "Age-dependent material properties of the porcine
cerebrum: effect on pediatric inertial head injury criteria." J Biomech 31(12): 1119-1126. Van Essen, D. C. and D. L. Dierker (2007). "Surface-Based and Probabilistic Atlases of Primate
Cerebral Cortex." Neuron 56(2): 209-225. Watanabe, H., F. Andersen, et al. (2001). "MR-Based Statistical Atlas of the Göttingen Minipig
Brain." NeuroImage 14(5): 1089-1096. Zhang, Y., M. Brady, et al. (2001). "Segmentation of brain MR images through a hidden Markov
random field model and the expectation-maximization algorithm." IEEE T Med Imaging 20(1): 45-57.